node-autotask 0.8.3 → 0.10.0

package/MCP_TOOL_CATALOG.json

@@ -1,7 +1,7 @@
 {
   "generatedBy": "scripts/build-tool-catalog.mjs",
-  "generatedAt": "2026-10-08T12:27:43.504Z",
-  "planHash": "e229200fc177d4428a3d73e75be413c91338876fdc1224bc3c9469788f96d975",
+  "generatedAt": "2026-10-11T12:48:06.777Z",
+  "planHash": "4c8d9f800e97c350acc681ade183f67474c044b674494eb78cd0f14a933dc1fc",
   "metaTools": [
     {
       "name": "autotask_catalog",
@@ -43,7 +43,7 @@
     {
       "name": "autotask_read",
       "title": "Read One Registry Operation (read mode)",
-      "description": "Route an effect=read registry operation by its exact name (never fuzzy, never a tool name): the 363 field/entity/UDF-info reads (getEntityInfo/getFieldInfo/getUdfInfo) have no dedicated tool and ride this one, and any read op is routable. Read-only by construction — every op it can route is effect=read (zero mutating requests), and a write or delete op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). Reads have no dry run (a read with dry_run: true is refused). Harness approval policy: approve every tool with readOnlyHint !== true (exactly 254 today); the 571 readOnlyHint: true tools — including this one — are safe ungated; the 55 destructiveHint: true tools (54 dedicated delete tools + autotask_delete) require confirm === the operation key. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. For query/count ops args[0] is a params object (filter/limit); call autotask_describe first for the exact arg shape and filter grammar. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
+      "description": "Route an effect=read registry operation by its exact name (never fuzzy, never a tool name): the 363 field/entity/UDF-info reads (getEntityInfo/getFieldInfo/getUdfInfo) have no dedicated tool and ride this one, and any read op is routable. For field discovery use operations.fieldCatalog (compact) instead of per-resource getFieldInfo (full dump). Read-only by construction — every op it can route is effect=read (zero mutating requests), and a write or delete op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). Reads have no dry run (a read with dry_run: true is refused). Harness approval policy: approve every tool with readOnlyHint !== true (exactly 254 today); the 600 readOnlyHint: true tools — including this one — are safe ungated; the 55 destructiveHint: true tools (54 dedicated delete tools + autotask_delete) require confirm === the operation key. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. For query/count ops args[0] is a params object (filter/limit); call autotask_describe first for the exact arg shape and filter grammar. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
       "input": {
         "operation": {
           "type": "string",
@@ -63,7 +63,7 @@
     {
       "name": "autotask_write",
       "title": "Write One Registry Operation (write mode)",
-      "description": "Route an effect=write registry operation by its exact name (never fuzzy, never a tool name), dry-run-first: the first call always travels dry_run: true (the pinned DryRunResult envelope, simulated: true — the mutating request is never issued); pass dry_run: false to apply. A read or delete op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). The 3 approval-gated writes (configuration_items.moveConfigurationItem, contacts.moveToCompany, resources.transferOwnership) additionally require confirm === the operation key, in BOTH dry-run and apply mode. Harness approval policy: this tool is one of the 254 tools that require approval (readOnlyHint !== true); the 571 readOnlyHint: true tools are safe ungated, and the 55 destructiveHint: true tools (54 dedicated delete tools + autotask_delete) require confirm === the operation key. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. For query/count ops args[0] is a params object (filter/limit); call autotask_describe first for the exact arg shape and filter grammar. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
+      "description": "Route an effect=write registry operation by its exact name (never fuzzy, never a tool name), dry-run-first: the first call always travels dry_run: true (the pinned DryRunResult envelope, simulated: true — the mutating request is never issued); pass dry_run: false to apply. A read or delete op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). The 3 approval-gated writes (configuration_items.moveConfigurationItem, contacts.moveToCompany, resources.transferOwnership) additionally require confirm === the operation key, in BOTH dry-run and apply mode. Harness approval policy: this tool is one of the 254 tools that require approval (readOnlyHint !== true); the 600 readOnlyHint: true tools are safe ungated, and the 55 destructiveHint: true tools (54 dedicated delete tools + autotask_delete) require confirm === the operation key. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. For query/count ops args[0] is a params object (filter/limit); call autotask_describe first for the exact arg shape and filter grammar. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
       "input": {
         "operation": {
           "type": "string",
@@ -93,7 +93,7 @@
     {
       "name": "autotask_delete",
       "title": "Delete One Registry Operation (delete mode)",
-      "description": "Route an effect=destructive (delete) registry operation by its exact name (never fuzzy, never a tool name), dry-run-first: the first call always travels dry_run: true (the pinned DryRunResult envelope, simulated: true); pass dry_run: false to apply. confirm === the operation key is REQUIRED for every delete (all 54), in BOTH dry-run and apply mode — the deliberate act is the call, not the delete. A read or write op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). Harness approval policy: this tool is one of the 55 destructiveHint: true tools that require approval + confirm; approve every tool with readOnlyHint !== true (exactly 254 today); the 571 readOnlyHint: true tools are safe ungated. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
+      "description": "Route an effect=destructive (delete) registry operation by its exact name (never fuzzy, never a tool name), dry-run-first: the first call always travels dry_run: true (the pinned DryRunResult envelope, simulated: true); pass dry_run: false to apply. confirm === the operation key is REQUIRED for every delete (all 54), in BOTH dry-run and apply mode — the deliberate act is the call, not the delete. A read or write op is refused BEFORE any request (cross-mode refusal; the message names the correct tool). Harness approval policy: this tool is one of the 55 destructiveHint: true tools that require approval + confirm; approve every tool with readOnlyHint !== true (exactly 254 today); the 600 readOnlyHint: true tools are safe ungated. Migration: autotask_invoke is REPLACED by autotask_read/autotask_write/autotask_delete — update harness config. The argument set is CLOSED at the top level: an unknown argument is refused before any request, with a typed error naming the key and the valid keys.",
       "input": {
         "operation": {
           "type": "string",
@@ -306,7 +306,7 @@
     {
       "operation": "appointments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Appointment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Appointment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "appointments.getFieldInfo",
       "effect": "read",
@@ -339,7 +339,7 @@
       "operation": "appointments.query",
       "tool": "autotask_appointment_query",
       "purpose": "Query Appointment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via appointments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "appointments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -457,7 +457,7 @@
     {
       "operation": "billing_codes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the BillingCode field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the BillingCode field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "billing_codes.getFieldInfo",
       "effect": "read",
@@ -490,7 +490,7 @@
       "operation": "billing_codes.query",
       "tool": "autotask_billingCode_query",
       "purpose": "Query BillingCode records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via billing_codes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "billing_codes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -554,6 +554,25 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "billing_items.customRange",
+      "tool": "autotask_billingItems_customRange",
+      "purpose": "Get billing items for an explicit itemDate range.",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.customRange",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "from",
+        "to"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "billing_items.get",
       "tool": "autotask_billingItems_get",
@@ -591,7 +610,7 @@
     {
       "operation": "billing_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the BillingItems field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the BillingItems field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "billing_items.getFieldInfo",
       "effect": "read",
@@ -620,11 +639,155 @@
       "reachReason": "no dedicated tool — the op is still discoverable and invocable by exact name",
       "boundedAlternative": null
     },
+    {
+      "operation": "billing_items.last14",
+      "tool": "autotask_billingItems_last14",
+      "purpose": "Get billing items for the last 14 days (rolling, itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.last14",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.last24h",
+      "tool": "autotask_billingItems_last24h",
+      "purpose": "Get billing items for the last 24 hours (rolling, itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.last24h",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.last30",
+      "tool": "autotask_billingItems_last30",
+      "purpose": "Get billing items for the last 30 days (rolling, itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.last30",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.last7",
+      "tool": "autotask_billingItems_last7",
+      "purpose": "Get billing items for the last 7 days (rolling, itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.last7",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.last90",
+      "tool": "autotask_billingItems_last90",
+      "purpose": "Get billing items for the last 90 days (rolling, itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.last90",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.lastFullFortnight",
+      "tool": "autotask_billingItems_lastFullFortnight",
+      "purpose": "Get billing items for the last complete fortnight (itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.lastFullFortnight",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.lastFullMonth",
+      "tool": "autotask_billingItems_lastFullMonth",
+      "purpose": "Get billing items for the last complete calendar month (itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.lastFullMonth",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.lastFullQuarter",
+      "tool": "autotask_billingItems_lastFullQuarter",
+      "purpose": "Get billing items for the last complete calendar quarter (itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.lastFullQuarter",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "billing_items.lastFullWeek",
+      "tool": "autotask_billingItems_lastFullWeek",
+      "purpose": "Get billing items for the last complete ISO week (itemDate).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.lastFullWeek",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "billing_items.query",
       "tool": "autotask_billingItems_query",
       "purpose": "Query BillingItems records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via billing_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "billing_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -654,6 +817,22 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "billing_items.today",
+      "tool": "autotask_billingItems_today",
+      "purpose": "Get billing items for today (itemDate = now, server date).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.today",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "billing_items.update",
       "tool": "autotask_billingItems_update",
@@ -673,6 +852,22 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "billing_items.yesterday",
+      "tool": "autotask_billingItems_yesterday",
+      "purpose": "Get billing items for yesterday (itemDate = previous server day).",
+      "preferredWhen": null,
+      "backingOperation": "billing_items.yesterday",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "change_request_links.count",
       "tool": "autotask_changeRequestLinks_count",
@@ -780,7 +975,7 @@
     {
       "operation": "change_request_links.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ChangeRequestLink field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ChangeRequestLink field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "change_request_links.getFieldInfo",
       "effect": "read",
@@ -813,7 +1008,7 @@
       "operation": "change_request_links.query",
       "tool": "autotask_changeRequestLinks_query",
       "purpose": "Query ChangeRequestLink records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via change_request_links.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "change_request_links.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -932,7 +1127,7 @@
     {
       "operation": "checklist_libraries.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ChecklistLibrary field catalog (types, picklists, required, maxLength) — reusable checklist definition.",
+      "purpose": "Get the ChecklistLibrary field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog); reusable checklist definition.",
       "preferredWhen": null,
       "backingOperation": "checklist_libraries.getFieldInfo",
       "effect": "read",
@@ -965,7 +1160,7 @@
       "operation": "checklist_libraries.query",
       "tool": "autotask_checklistLibrary_query",
       "purpose": "Query ChecklistLibrary records (reusable checklist definitions) with a JSON filter. Not Tasks.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via checklist_libraries.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "checklist_libraries.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1124,7 +1319,7 @@
     {
       "operation": "checklist_library_checklist_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ChecklistLibraryChecklistItem field catalog (types, picklists, required, maxLength) — reusable library checklist step.",
+      "purpose": "Get the ChecklistLibraryChecklistItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog); reusable library checklist step.",
       "preferredWhen": null,
       "backingOperation": "checklist_library_checklist_items.getFieldInfo",
       "effect": "read",
@@ -1157,7 +1352,7 @@
       "operation": "checklist_library_checklist_items.query",
       "tool": "autotask_checklistLibraryChecklistItem_query",
       "purpose": "Query ChecklistLibraryChecklistItem records (reusable library checklist steps) with a JSON filter. Not Tasks; narrow to a library with the checklistLibraryID filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via checklist_library_checklist_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "checklist_library_checklist_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1260,7 +1455,7 @@
     {
       "operation": "classification_icons.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ClassificationIcon field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ClassificationIcon field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "classification_icons.getFieldInfo",
       "effect": "read",
@@ -1293,7 +1488,7 @@
       "operation": "classification_icons.query",
       "tool": "autotask_classificationIcon_query",
       "purpose": "Query ClassificationIcon records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via classification_icons.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "classification_icons.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1412,7 +1607,7 @@
     {
       "operation": "configuration_item_categories.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemCategory field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemCategory field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_categories.getFieldInfo",
       "effect": "read",
@@ -1445,7 +1640,7 @@
       "operation": "configuration_item_categories.query",
       "tool": "autotask_configurationItemCategories_query",
       "purpose": "Query ConfigurationItemCategory records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_categories.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_categories.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1604,7 +1799,7 @@
     {
       "operation": "configuration_item_category_udf_associations.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemCategoryUdfAssociation field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemCategoryUdfAssociation field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_category_udf_associations.getFieldInfo",
       "effect": "read",
@@ -1637,7 +1832,7 @@
       "operation": "configuration_item_category_udf_associations.query",
       "tool": "autotask_configurationItemCategoryUdfAssociation_query",
       "purpose": "Query ConfigurationItemCategoryUdfAssociation records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_category_udf_associations.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_category_udf_associations.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1776,7 +1971,7 @@
     {
       "operation": "configuration_item_types.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemType field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemType field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_types.getFieldInfo",
       "effect": "read",
@@ -1809,7 +2004,7 @@
       "operation": "configuration_item_types.query",
       "tool": "autotask_configurationItemTypes_query",
       "purpose": "Query ConfigurationItemType records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_types.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_types.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -1965,7 +2160,7 @@
     {
       "operation": "contact_group_contacts.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContactGroupContacts field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContactGroupContacts field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contact_group_contacts.getFieldInfo",
       "effect": "read",
@@ -1998,7 +2193,7 @@
       "operation": "contact_group_contacts.query",
       "tool": "autotask_contactGroupContacts_query",
       "purpose": "Query ContactGroupContacts records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contact_group_contacts.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contact_group_contacts.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2117,7 +2312,7 @@
     {
       "operation": "contact_groups.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContactGroups field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContactGroups field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contact_groups.getFieldInfo",
       "effect": "read",
@@ -2150,7 +2345,7 @@
       "operation": "contact_groups.query",
       "tool": "autotask_contactGroups_query",
       "purpose": "Query ContactGroups records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contact_groups.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contact_groups.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2309,7 +2504,7 @@
     {
       "operation": "contract_exclusion_set_excluded_roles.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractExclusionSetExcludedRoles field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractExclusionSetExcludedRoles field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_exclusion_set_excluded_roles.getFieldInfo",
       "effect": "read",
@@ -2342,7 +2537,7 @@
       "operation": "contract_exclusion_set_excluded_roles.query",
       "tool": "autotask_contractExclusionSetExcludedRole_query",
       "purpose": "Query ContractExclusionSetExcludedRoles records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_exclusion_set_excluded_roles.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_exclusion_set_excluded_roles.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2464,7 +2659,7 @@
     {
       "operation": "contract_exclusion_set_excluded_work_types.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractExclusionSetExcludedWorkTypes field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractExclusionSetExcludedWorkTypes field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_exclusion_set_excluded_work_types.getFieldInfo",
       "effect": "read",
@@ -2497,7 +2692,7 @@
       "operation": "contract_exclusion_set_excluded_work_types.query",
       "tool": "autotask_contractExclusionSetExcludedWorkType_query",
       "purpose": "Query ContractExclusionSetExcludedWorkTypes records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_exclusion_set_excluded_work_types.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_exclusion_set_excluded_work_types.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2616,7 +2811,7 @@
     {
       "operation": "contract_exclusion_sets.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractExclusionSets field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractExclusionSets field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_exclusion_sets.getFieldInfo",
       "effect": "read",
@@ -2649,7 +2844,7 @@
       "operation": "contract_exclusion_sets.query",
       "tool": "autotask_contractExclusionSets_query",
       "purpose": "Query ContractExclusionSets records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_exclusion_sets.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_exclusion_sets.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2769,7 +2964,7 @@
     {
       "operation": "countries.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Country field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Country field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "countries.getFieldInfo",
       "effect": "read",
@@ -2802,7 +2997,7 @@
       "operation": "countries.query",
       "tool": "autotask_country_query",
       "purpose": "Query Country records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via countries.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "countries.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -2940,7 +3135,7 @@
     {
       "operation": "departments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Department field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Department field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "departments.getFieldInfo",
       "effect": "read",
@@ -2973,7 +3168,7 @@
       "operation": "departments.query",
       "tool": "autotask_department_query",
       "purpose": "Query Department records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via departments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "departments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -3111,7 +3306,7 @@
     {
       "operation": "domain_registrars.getFieldInfo",
       "tool": null,
-      "purpose": "Get the DomainRegistrar field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the DomainRegistrar field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "domain_registrars.getFieldInfo",
       "effect": "read",
@@ -3144,7 +3339,7 @@
       "operation": "domain_registrars.query",
       "tool": "autotask_DomainRegistrar_query",
       "purpose": "Query DomainRegistrar records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via domain_registrars.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "domain_registrars.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -3338,7 +3533,7 @@
     {
       "operation": "expense_item_attachments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ExpenseItemAttachment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ExpenseItemAttachment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "expense_item_attachments.getFieldInfo",
       "effect": "read",
@@ -3371,7 +3566,7 @@
       "operation": "expense_item_attachments.query",
       "tool": "autotask_expenseItemAttachment_query",
       "purpose": "Query ExpenseItemAttachment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via expense_item_attachments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "expense_item_attachments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -3492,7 +3687,7 @@
     {
       "operation": "expense_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ExpenseItem field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ExpenseItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "expense_items.getFieldInfo",
       "effect": "read",
@@ -3525,7 +3720,7 @@
       "operation": "expense_items.query",
       "tool": "autotask_expenseItem_query",
       "purpose": "Query ExpenseItem records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via expense_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "expense_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -3646,7 +3841,7 @@
     {
       "operation": "expense_reports.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ExpenseReport field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ExpenseReport field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "expense_reports.getFieldInfo",
       "effect": "read",
@@ -3679,7 +3874,7 @@
       "operation": "expense_reports.query",
       "tool": "autotask_expenseReport_query",
       "purpose": "Query ExpenseReport records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via expense_reports.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "expense_reports.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -3853,7 +4048,7 @@
     {
       "operation": "holiday_sets.getFieldInfo",
       "tool": null,
-      "purpose": "Get the HolidaySet field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the HolidaySet field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "holiday_sets.getFieldInfo",
       "effect": "read",
@@ -3886,7 +4081,7 @@
       "operation": "holiday_sets.query",
       "tool": "autotask_holidaySet_query",
       "purpose": "Query HolidaySet records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via holiday_sets.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "holiday_sets.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4046,7 +4241,7 @@
     {
       "operation": "holidays.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Holiday field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Holiday field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "holidays.getFieldInfo",
       "effect": "read",
@@ -4079,7 +4274,7 @@
       "operation": "holidays.query",
       "tool": "autotask_holiday_query",
       "purpose": "Query Holiday records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via holidays.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "holidays.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4218,7 +4413,7 @@
     {
       "operation": "inventory_products.getFieldInfo",
       "tool": null,
-      "purpose": "Get the InventoryProduct field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the InventoryProduct field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "inventory_products.getFieldInfo",
       "effect": "read",
@@ -4251,7 +4446,7 @@
       "operation": "inventory_products.query",
       "tool": "autotask_inventoryProduct_query",
       "purpose": "Query InventoryProduct records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via inventory_products.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "inventory_products.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4426,7 +4621,7 @@
     {
       "operation": "notification_history.getFieldInfo",
       "tool": null,
-      "purpose": "Get the NotificationHistory field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the NotificationHistory field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "notification_history.getFieldInfo",
       "effect": "read",
@@ -4459,7 +4654,7 @@
       "operation": "notification_history.query",
       "tool": "autotask_notificationHistory_query",
       "purpose": "Query NotificationHistory records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via notification_history.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "notification_history.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4542,7 +4737,7 @@
     {
       "operation": "opportunity_categories.getFieldInfo",
       "tool": null,
-      "purpose": "Get the OpportunityCategory field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the OpportunityCategory field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "opportunity_categories.getFieldInfo",
       "effect": "read",
@@ -4575,7 +4770,7 @@
       "operation": "opportunity_categories.query",
       "tool": "autotask_opportunityCategories_query",
       "purpose": "Query OpportunityCategory records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via opportunity_categories.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "opportunity_categories.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4715,7 +4910,7 @@
     {
       "operation": "product_vendors.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ProductVendors field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ProductVendors field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "product_vendors.getFieldInfo",
       "effect": "read",
@@ -4748,7 +4943,7 @@
       "operation": "product_vendors.query",
       "tool": "autotask_productVendor_query",
       "purpose": "Query ProductVendors records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via product_vendors.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "product_vendors.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -4869,7 +5064,7 @@
     {
       "operation": "products.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Product field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Product field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "products.getFieldInfo",
       "effect": "read",
@@ -4902,7 +5097,7 @@
       "operation": "products.query",
       "tool": "autotask_product_query",
       "purpose": "Query Product records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via products.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "products.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5022,7 +5217,7 @@
     {
       "operation": "purchase_approvals.getFieldInfo",
       "tool": null,
-      "purpose": "Get the PurchaseApprovals field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the PurchaseApprovals field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "purchase_approvals.getFieldInfo",
       "effect": "read",
@@ -5055,7 +5250,7 @@
       "operation": "purchase_approvals.query",
       "tool": "autotask_purchaseApprovals_query",
       "purpose": "Query PurchaseApprovals records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via purchase_approvals.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "purchase_approvals.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5177,7 +5372,7 @@
     {
       "operation": "purchase_order_item_receiving.getFieldInfo",
       "tool": null,
-      "purpose": "Get the PurchaseOrderItemReceiving field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the PurchaseOrderItemReceiving field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "purchase_order_item_receiving.getFieldInfo",
       "effect": "read",
@@ -5210,7 +5405,7 @@
       "operation": "purchase_order_item_receiving.query",
       "tool": "autotask_purchaseOrderItemReceiving_query",
       "purpose": "Query PurchaseOrderItemReceiving records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via purchase_order_item_receiving.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "purchase_order_item_receiving.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5313,7 +5508,7 @@
     {
       "operation": "purchase_order_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the PurchaseOrderItems field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the PurchaseOrderItems field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "purchase_order_items.getFieldInfo",
       "effect": "read",
@@ -5346,7 +5541,7 @@
       "operation": "purchase_order_items.query",
       "tool": "autotask_purchaseOrderItems_query",
       "purpose": "Query PurchaseOrderItems records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via purchase_order_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "purchase_order_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5467,7 +5662,7 @@
     {
       "operation": "purchase_orders.getFieldInfo",
       "tool": null,
-      "purpose": "Get the PurchaseOrders field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the PurchaseOrders field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "purchase_orders.getFieldInfo",
       "effect": "read",
@@ -5500,7 +5695,7 @@
       "operation": "purchase_orders.query",
       "tool": "autotask_purchaseOrders_query",
       "purpose": "Query PurchaseOrders records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via purchase_orders.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "purchase_orders.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5602,7 +5797,7 @@
     {
       "operation": "quote_templates.getFieldInfo",
       "tool": null,
-      "purpose": "Get the QuoteTemplate field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the QuoteTemplate field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "quote_templates.getFieldInfo",
       "effect": "read",
@@ -5635,7 +5830,7 @@
       "operation": "quote_templates.query",
       "tool": "autotask_quoteTemplate_query",
       "purpose": "Query QuoteTemplate records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via quote_templates.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "quote_templates.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5756,7 +5951,7 @@
     {
       "operation": "resource_role_queues.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceRoleQueue field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceRoleQueue field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_role_queues.getFieldInfo",
       "effect": "read",
@@ -5789,7 +5984,7 @@
       "operation": "resource_role_queues.query",
       "tool": "autotask_resourceRoleQueue_query",
       "purpose": "Query ResourceRoleQueue records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via resource_role_queues.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "resource_role_queues.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -5893,7 +6088,7 @@
     {
       "operation": "resource_roles.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceRole field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceRole field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_roles.getFieldInfo",
       "effect": "read",
@@ -5926,7 +6121,7 @@
       "operation": "resource_roles.query",
       "tool": "autotask_resourceRole_query",
       "purpose": "Query ResourceRole records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via resource_roles.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "resource_roles.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -6010,7 +6205,7 @@
     {
       "operation": "resource_skills.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceSkill field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceSkill field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_skills.getFieldInfo",
       "effect": "read",
@@ -6043,7 +6238,7 @@
       "operation": "resource_skills.query",
       "tool": "autotask_resourceSkill_query",
       "purpose": "Query ResourceSkill records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via resource_skills.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "resource_skills.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -6130,7 +6325,7 @@
     {
       "operation": "resource_time_off_additional.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceTimeOffAdditional field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceTimeOffAdditional field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_time_off_additional.getFieldInfo",
       "effect": "read",
@@ -6251,7 +6446,7 @@
     {
       "operation": "resource_time_off_approvers.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceTimeOffApprover field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceTimeOffApprover field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_time_off_approvers.getFieldInfo",
       "effect": "read",
@@ -6284,7 +6479,7 @@
       "operation": "resource_time_off_approvers.query",
       "tool": "autotask_resourceTimeOffApprover_query",
       "purpose": "Query ResourceTimeOffApprover records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via resource_time_off_approvers.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "resource_time_off_approvers.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -6389,7 +6584,7 @@
     {
       "operation": "resource_time_off_balances.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ResourceTimeOffBalance field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ResourceTimeOffBalance field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resource_time_off_balances.getFieldInfo",
       "effect": "read",
@@ -6489,7 +6684,7 @@
     {
       "operation": "resources.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Resource field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Resource field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "resources.getFieldInfo",
       "effect": "read",
@@ -6522,7 +6717,7 @@
       "operation": "resources.query",
       "tool": "autotask_resource_query",
       "purpose": "Query Resource records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via resources.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "resources.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -6679,7 +6874,7 @@
     {
       "operation": "roles.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Role field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Role field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "roles.getFieldInfo",
       "effect": "read",
@@ -6712,7 +6907,7 @@
       "operation": "roles.query",
       "tool": "autotask_role_query",
       "purpose": "Query Role records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via roles.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "roles.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -6871,7 +7066,7 @@
     {
       "operation": "service_call_task_resources.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceCallTaskResource field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceCallTaskResource field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_call_task_resources.getFieldInfo",
       "effect": "read",
@@ -6904,7 +7099,7 @@
       "operation": "service_call_task_resources.query",
       "tool": "autotask_serviceCallTaskResource_query",
       "purpose": "Query ServiceCallTaskResource records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_call_task_resources.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_call_task_resources.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7026,7 +7221,7 @@
     {
       "operation": "service_call_tasks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceCallTask field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceCallTask field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_call_tasks.getFieldInfo",
       "effect": "read",
@@ -7059,7 +7254,7 @@
       "operation": "service_call_tasks.query",
       "tool": "autotask_serviceCallTask_query",
       "purpose": "Query ServiceCallTask records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_call_tasks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_call_tasks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7181,7 +7376,7 @@
     {
       "operation": "service_call_ticket_resources.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceCallTicketResource field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceCallTicketResource field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_call_ticket_resources.getFieldInfo",
       "effect": "read",
@@ -7214,7 +7409,7 @@
       "operation": "service_call_ticket_resources.query",
       "tool": "autotask_serviceCallTicketResource_query",
       "purpose": "Query ServiceCallTicketResource records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_call_ticket_resources.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_call_ticket_resources.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7336,7 +7531,7 @@
     {
       "operation": "service_call_tickets.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceCallTicket field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceCallTicket field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_call_tickets.getFieldInfo",
       "effect": "read",
@@ -7369,7 +7564,7 @@
       "operation": "service_call_tickets.query",
       "tool": "autotask_serviceCallTicket_query",
       "purpose": "Query ServiceCallTicket records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_call_tickets.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_call_tickets.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7488,7 +7683,7 @@
     {
       "operation": "service_calls.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceCall field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceCall field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_calls.getFieldInfo",
       "effect": "read",
@@ -7521,7 +7716,7 @@
       "operation": "service_calls.query",
       "tool": "autotask_serviceCall_query",
       "purpose": "Query ServiceCall records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_calls.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_calls.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7623,7 +7818,7 @@
     {
       "operation": "service_level_agreement_results.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ServiceLevelAgreementResults field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ServiceLevelAgreementResults field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "service_level_agreement_results.getFieldInfo",
       "effect": "read",
@@ -7656,7 +7851,7 @@
       "operation": "service_level_agreement_results.query",
       "tool": "autotask_serviceLevelAgreementResults_query",
       "purpose": "Query ServiceLevelAgreementResults records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via service_level_agreement_results.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "service_level_agreement_results.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7757,7 +7952,7 @@
     {
       "operation": "services.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Service field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Service field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "services.getFieldInfo",
       "effect": "read",
@@ -7790,7 +7985,7 @@
       "operation": "services.query",
       "tool": "autotask_service_query",
       "purpose": "Query Service records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via services.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "services.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -7910,7 +8105,7 @@
     {
       "operation": "skills.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Skill field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Skill field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "skills.getFieldInfo",
       "effect": "read",
@@ -7943,7 +8138,7 @@
       "operation": "skills.query",
       "tool": "autotask_skill_query",
       "purpose": "Query Skill records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via skills.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "skills.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8044,7 +8239,7 @@
     {
       "operation": "subscription_periods.getFieldInfo",
       "tool": null,
-      "purpose": "Get the SubscriptionPeriod field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the SubscriptionPeriod field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "subscription_periods.getFieldInfo",
       "effect": "read",
@@ -8077,7 +8272,7 @@
       "operation": "subscription_periods.query",
       "tool": "autotask_subscriptionPeriod_query",
       "purpose": "Query SubscriptionPeriod records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via subscription_periods.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "subscription_periods.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8160,7 +8355,7 @@
     {
       "operation": "survey_results.getFieldInfo",
       "tool": null,
-      "purpose": "Get the SurveyResults field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the SurveyResults field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "survey_results.getFieldInfo",
       "effect": "read",
@@ -8193,7 +8388,7 @@
       "operation": "survey_results.query",
       "tool": "autotask_surveyResults_query",
       "purpose": "Query SurveyResults records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via survey_results.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "survey_results.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8276,7 +8471,7 @@
     {
       "operation": "surveys.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Survey field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Survey field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "surveys.getFieldInfo",
       "effect": "read",
@@ -8309,7 +8504,7 @@
       "operation": "surveys.query",
       "tool": "autotask_survey_query",
       "purpose": "Query Survey records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via surveys.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "surveys.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8449,7 +8644,7 @@
     {
       "operation": "tag_aliases.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TagAlias field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TagAlias field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "tag_aliases.getFieldInfo",
       "effect": "read",
@@ -8482,7 +8677,7 @@
       "operation": "tag_aliases.query",
       "tool": "autotask_tagAlias_query",
       "purpose": "Query TagAlias records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via tag_aliases.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "tag_aliases.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8601,7 +8796,7 @@
     {
       "operation": "tag_groups.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TagGroup field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TagGroup field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "tag_groups.getFieldInfo",
       "effect": "read",
@@ -8634,7 +8829,7 @@
       "operation": "tag_groups.query",
       "tool": "autotask_tagGroup_query",
       "purpose": "Query TagGroup records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via tag_groups.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "tag_groups.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8772,7 +8967,7 @@
     {
       "operation": "tags.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Tag field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Tag field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "tags.getFieldInfo",
       "effect": "read",
@@ -8805,7 +9000,7 @@
       "operation": "tags.query",
       "tool": "autotask_tag_query",
       "purpose": "Query Tag records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via tags.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "tags.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -8907,7 +9102,7 @@
     {
       "operation": "ticket_categories.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketCategory field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketCategory field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_categories.getFieldInfo",
       "effect": "read",
@@ -8940,7 +9135,7 @@
       "operation": "ticket_categories.query",
       "tool": "autotask_ticketCategories_query",
       "purpose": "Query TicketCategory records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_categories.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_categories.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -9061,7 +9256,7 @@
     {
       "operation": "ticket_category_field_defaults.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketCategoryFieldDefault field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketCategoryFieldDefault field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_category_field_defaults.getFieldInfo",
       "effect": "read",
@@ -9094,7 +9289,7 @@
       "operation": "ticket_category_field_defaults.query",
       "tool": "autotask_ticketCategoryFieldDefaults_query",
       "purpose": "Query TicketCategoryFieldDefault records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_category_field_defaults.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_category_field_defaults.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -9177,7 +9372,7 @@
     {
       "operation": "ticket_history.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketHistory field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketHistory field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_history.getFieldInfo",
       "effect": "read",
@@ -9210,7 +9405,7 @@
       "operation": "ticket_history.query",
       "tool": "autotask_ticketHistory_query",
       "purpose": "Query TicketHistory records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_history.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_history.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -9351,7 +9546,7 @@
     {
       "operation": "ticket_note_attachments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketNoteAttachment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketNoteAttachment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_note_attachments.getFieldInfo",
       "effect": "read",
@@ -9384,7 +9579,7 @@
       "operation": "ticket_note_attachments.query",
       "tool": "autotask_ticketNoteAttachment_query",
       "purpose": "Query TicketNoteAttachment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_note_attachments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_note_attachments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -9503,7 +9698,7 @@
     {
       "operation": "ticket_note_webhooks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketNoteWebhook field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketNoteWebhook field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_note_webhooks.getFieldInfo",
       "effect": "read",
@@ -9536,7 +9731,7 @@
       "operation": "ticket_note_webhooks.query",
       "tool": "autotask_ticketNoteWebhook_query",
       "purpose": "Query TicketNoteWebhook records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_note_webhooks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_note_webhooks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -9693,6 +9888,25 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "time_entries.dateWorkedWindow",
+      "tool": "autotask_timeEntry_date_worked_window",
+      "purpose": "The clearer-named alias for time_entries.postedWithin (issue #66): the dateWorked window read with the posted/unposted split, counted in one pass — same wire, same result (incl. truncated).",
+      "preferredWhen": null,
+      "backingOperation": "time_entries.dateWorkedWindow",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "from",
+        "to"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "time_entries.delete",
       "tool": "autotask_timeEntry_delete",
@@ -9748,7 +9962,7 @@
     {
       "operation": "time_entries.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TimeEntry field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TimeEntry field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "time_entries.getFieldInfo",
       "effect": "read",
@@ -9956,7 +10170,7 @@
     {
       "operation": "time_entries.postedWithin",
       "tool": "autotask_timeEntry_posted_within",
-      "purpose": "Get time entries worked within a UTC date window (inclusive) with the posted/unposted split, counted in one pass.",
+      "purpose": "Get time entries worked within an inclusive dateWorked window with the posted/unposted split, counted in one pass (the window is ON the dateWorked field — the posted narrowing is a server-side postedDate exist|notExist predicate; the counts cover the returned rows only, truncated flags the capped window).",
       "preferredWhen": null,
       "backingOperation": "time_entries.postedWithin",
       "effect": "read",
@@ -9976,7 +10190,7 @@
       "operation": "time_entries.query",
       "tool": "autotask_timeEntry_query",
       "purpose": "Query TimeEntry records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via time_entries.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "time_entries.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -10168,7 +10382,7 @@
     {
       "operation": "time_entry_attachments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TimeEntryAttachment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TimeEntryAttachment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "time_entry_attachments.getFieldInfo",
       "effect": "read",
@@ -10201,7 +10415,7 @@
       "operation": "time_entry_attachments.query",
       "tool": "autotask_timeEntryAttachment_query",
       "purpose": "Query TimeEntryAttachment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via time_entry_attachments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "time_entry_attachments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -10302,7 +10516,7 @@
     {
       "operation": "time_off_requests.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TimeOffRequest field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TimeOffRequest field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "time_off_requests.getFieldInfo",
       "effect": "read",
@@ -10335,7 +10549,7 @@
       "operation": "time_off_requests.query",
       "tool": "autotask_timeOffRequest_query",
       "purpose": "Query TimeOffRequest records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via time_off_requests.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "time_off_requests.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -10523,7 +10737,7 @@
     {
       "operation": "companies.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Company field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Company field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "companies.getFieldInfo",
       "effect": "read",
@@ -10556,7 +10770,7 @@
       "operation": "companies.query",
       "tool": "autotask_company_query",
       "purpose": "Query Company records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via companies.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "companies.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -10572,7 +10786,7 @@
       "operation": "companies.search",
       "tool": "autotask_company_search",
       "purpose": "Search company records by text (contains on the searchable fields).",
-      "preferredWhen": null,
+      "preferredWhen": "Fuzzy names, abbreviations and family lookups (the contains channel — exact match is the resource's resolve; a resolve miss or the RESOLUTION_AMBIGUOUS / RESOLUTION_TRUNCATED errors name this op via the error's suggestedAction). With includeFamily: true each match carries its parentID + childIDs (one bounded query per match — blind-surface scoped, see the family block's blindNote).",
       "backingOperation": "companies.search",
       "effect": "read",
       "dryRunFirst": false,
@@ -10622,6 +10836,24 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "companies.unbilledWork",
+      "tool": "autotask_company_unbilled_work",
+      "purpose": "One call for a company's unbilled work: its posted billing items (per source), the time behind them (joined through the company's tickets and tasks — time entries carry no companyID), the per-ticket rollup, the unposted anti-join section, ex-tax totals + hours, and the billing-vs-time warnings",
+      "preferredWhen": "For allocation-style questions about one company's posted/uninvoiced work — replaces the multi-call billing×time×ticket hand-join; companies.fullProfile for the general company picture; operations.aggregate for cross-resource existence/totals",
+      "backingOperation": "companies.unbilledWork",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "id"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "companies.update",
       "tool": "autotask_company_update",
@@ -10752,7 +10984,7 @@
     {
       "operation": "company_alerts.getFieldInfo",
       "tool": null,
-      "purpose": "Get the CompanyAlert field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the CompanyAlert field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "company_alerts.getFieldInfo",
       "effect": "read",
@@ -10785,7 +11017,7 @@
       "operation": "company_alerts.query",
       "tool": "autotask_companyAlert_query",
       "purpose": "Query CompanyAlert records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via company_alerts.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "company_alerts.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -10927,7 +11159,7 @@
     {
       "operation": "company_locations.getFieldInfo",
       "tool": null,
-      "purpose": "Get the CompanyLocation field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the CompanyLocation field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "company_locations.getFieldInfo",
       "effect": "read",
@@ -10960,7 +11192,7 @@
       "operation": "company_locations.query",
       "tool": "autotask_companyLocation_query",
       "purpose": "Query CompanyLocation records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via company_locations.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "company_locations.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11101,7 +11333,7 @@
     {
       "operation": "company_notes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the CompanyNote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the CompanyNote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "company_notes.getFieldInfo",
       "effect": "read",
@@ -11134,7 +11366,7 @@
       "operation": "company_notes.query",
       "tool": "autotask_companyNote_query",
       "purpose": "Query CompanyNote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via company_notes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "company_notes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11256,7 +11488,7 @@
     {
       "operation": "company_site_configurations.getFieldInfo",
       "tool": null,
-      "purpose": "Get the CompanySiteConfigurations field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the CompanySiteConfigurations field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "company_site_configurations.getFieldInfo",
       "effect": "read",
@@ -11289,7 +11521,7 @@
       "operation": "company_site_configurations.query",
       "tool": "autotask_companySiteConfigurations_query",
       "purpose": "Query CompanySiteConfigurations records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via company_site_configurations.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "company_site_configurations.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11428,7 +11660,7 @@
     {
       "operation": "company_webhooks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the CompanyWebhook field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the CompanyWebhook field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "company_webhooks.getFieldInfo",
       "effect": "read",
@@ -11461,7 +11693,7 @@
       "operation": "company_webhooks.query",
       "tool": "autotask_companyWebhook_query",
       "purpose": "Query CompanyWebhook records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via company_webhooks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "company_webhooks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11620,7 +11852,7 @@
     {
       "operation": "configuration_item_billing_product_associations.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemBillingProductAssociations field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemBillingProductAssociations field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_billing_product_associations.getFieldInfo",
       "effect": "read",
@@ -11653,7 +11885,7 @@
       "operation": "configuration_item_billing_product_associations.query",
       "tool": "autotask_configurationItemBillingProductAssociation_query",
       "purpose": "Query ConfigurationItemBillingProductAssociations records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_billing_product_associations.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_billing_product_associations.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11776,7 +12008,7 @@
     {
       "operation": "configuration_item_dns_records.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemDnsRecord field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemDnsRecord field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_dns_records.getFieldInfo",
       "effect": "read",
@@ -11809,7 +12041,7 @@
       "operation": "configuration_item_dns_records.query",
       "tool": "autotask_configurationItemDnsRecord_query",
       "purpose": "Query ConfigurationItemDnsRecord records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_dns_records.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_dns_records.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -11912,7 +12144,7 @@
     {
       "operation": "configuration_item_notes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemNote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemNote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_notes.getFieldInfo",
       "effect": "read",
@@ -11945,7 +12177,7 @@
       "operation": "configuration_item_notes.query",
       "tool": "autotask_configurationItemNote_query",
       "purpose": "Query ConfigurationItemNote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_notes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_notes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -12087,7 +12319,7 @@
     {
       "operation": "configuration_item_related_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemRelatedItem field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemRelatedItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_related_items.getFieldInfo",
       "effect": "read",
@@ -12120,7 +12352,7 @@
       "operation": "configuration_item_related_items.query",
       "tool": "autotask_configurationItemRelatedItem_query",
       "purpose": "Query ConfigurationItemRelatedItem records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_related_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_related_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -12223,7 +12455,7 @@
     {
       "operation": "configuration_item_ssl_subject_alternative_names.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemSslSubjectAlternativeName field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemSslSubjectAlternativeName field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_ssl_subject_alternative_names.getFieldInfo",
       "effect": "read",
@@ -12256,7 +12488,7 @@
       "operation": "configuration_item_ssl_subject_alternative_names.query",
       "tool": "autotask_configurationItemSslSubjectAlternativeName_query",
       "purpose": "Query ConfigurationItemSslSubjectAlternativeName records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_ssl_subject_alternative_names.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_ssl_subject_alternative_names.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -12413,7 +12645,7 @@
     {
       "operation": "configuration_item_webhooks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItemWebhook field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItemWebhook field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_item_webhooks.getFieldInfo",
       "effect": "read",
@@ -12446,7 +12678,7 @@
       "operation": "configuration_item_webhooks.query",
       "tool": "autotask_configurationItemWebhook_query",
       "purpose": "Query ConfigurationItemWebhook records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_item_webhooks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_item_webhooks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -12638,7 +12870,7 @@
     {
       "operation": "configuration_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ConfigurationItem field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ConfigurationItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "configuration_items.getFieldInfo",
       "effect": "read",
@@ -12708,7 +12940,7 @@
       "operation": "configuration_items.query",
       "tool": "autotask_configurationItems_query",
       "purpose": "Query ConfigurationItem records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via configuration_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "configuration_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -12846,7 +13078,7 @@
     {
       "operation": "contact_webhooks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContactWebhook field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContactWebhook field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contact_webhooks.getFieldInfo",
       "effect": "read",
@@ -12879,7 +13111,7 @@
       "operation": "contact_webhooks.query",
       "tool": "autotask_contactWebhook_query",
       "purpose": "Query ContactWebhook records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contact_webhooks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contact_webhooks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13074,7 +13306,7 @@
     {
       "operation": "contacts.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Contact field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Contact field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contacts.getFieldInfo",
       "effect": "read",
@@ -13126,7 +13358,7 @@
       "operation": "contacts.query",
       "tool": "autotask_contact_query",
       "purpose": "Query Contact records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contacts.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contacts.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13287,7 +13519,7 @@
     {
       "operation": "contract_billing_rules.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractBillingRule field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractBillingRule field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_billing_rules.getFieldInfo",
       "effect": "read",
@@ -13320,7 +13552,7 @@
       "operation": "contract_billing_rules.query",
       "tool": "autotask_contractBillingRule_query",
       "purpose": "Query ContractBillingRule records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_billing_rules.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_billing_rules.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13443,7 +13675,7 @@
     {
       "operation": "contract_block_hour_factors.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractBlockHourFactor field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractBlockHourFactor field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_block_hour_factors.getFieldInfo",
       "effect": "read",
@@ -13476,7 +13708,7 @@
       "operation": "contract_block_hour_factors.query",
       "tool": "autotask_contractBlockHourFactor_query",
       "purpose": "Query ContractBlockHourFactor records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_block_hour_factors.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_block_hour_factors.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13599,7 +13831,7 @@
     {
       "operation": "contract_blocks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractBlock field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractBlock field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_blocks.getFieldInfo",
       "effect": "read",
@@ -13632,7 +13864,7 @@
       "operation": "contract_blocks.query",
       "tool": "autotask_contractBlock_query",
       "purpose": "Query ContractBlock records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_blocks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_blocks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13793,7 +14025,7 @@
     {
       "operation": "contract_charges.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractCharge field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractCharge field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_charges.getFieldInfo",
       "effect": "read",
@@ -13826,7 +14058,7 @@
       "operation": "contract_charges.query",
       "tool": "autotask_contractCharge_query",
       "purpose": "Query ContractCharge records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_charges.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_charges.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -13986,7 +14218,7 @@
     {
       "operation": "contract_exclusion_billing_codes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractExclusionBillingCode field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractExclusionBillingCode field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_exclusion_billing_codes.getFieldInfo",
       "effect": "read",
@@ -14019,7 +14251,7 @@
       "operation": "contract_exclusion_billing_codes.query",
       "tool": "autotask_contractExclusionBillingCode_query",
       "purpose": "Query ContractExclusionBillingCode records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_exclusion_billing_codes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_exclusion_billing_codes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14141,7 +14373,7 @@
     {
       "operation": "contract_exclusion_roles.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractExclusionRoles field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractExclusionRoles field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_exclusion_roles.getFieldInfo",
       "effect": "read",
@@ -14174,7 +14406,7 @@
       "operation": "contract_exclusion_roles.query",
       "tool": "autotask_contractExclusionRoles_query",
       "purpose": "Query ContractExclusionRoles records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_exclusion_roles.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_exclusion_roles.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14277,7 +14509,7 @@
     {
       "operation": "contract_milestones.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractMilestone field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractMilestone field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_milestones.getFieldInfo",
       "effect": "read",
@@ -14310,7 +14542,7 @@
       "operation": "contract_milestones.query",
       "tool": "autotask_contractMilestone_query",
       "purpose": "Query ContractMilestone records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_milestones.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_milestones.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14433,7 +14665,7 @@
     {
       "operation": "contract_notes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractNote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractNote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_notes.getFieldInfo",
       "effect": "read",
@@ -14466,7 +14698,7 @@
       "operation": "contract_notes.query",
       "tool": "autotask_contractNote_query",
       "purpose": "Query ContractNote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_notes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_notes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14589,7 +14821,7 @@
     {
       "operation": "contract_rates.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractRate field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractRate field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_rates.getFieldInfo",
       "effect": "read",
@@ -14622,7 +14854,7 @@
       "operation": "contract_rates.query",
       "tool": "autotask_contractRate_query",
       "purpose": "Query ContractRate records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_rates.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_rates.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14745,7 +14977,7 @@
     {
       "operation": "contract_retainers.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractRetainers field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractRetainers field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_retainers.getFieldInfo",
       "effect": "read",
@@ -14778,7 +15010,7 @@
       "operation": "contract_retainers.query",
       "tool": "autotask_contractRetainer_query",
       "purpose": "Query ContractRetainers records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_retainers.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_retainers.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -14901,7 +15133,7 @@
     {
       "operation": "contract_role_costs.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractRoleCosts field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractRoleCosts field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_role_costs.getFieldInfo",
       "effect": "read",
@@ -14934,7 +15166,7 @@
       "operation": "contract_role_costs.query",
       "tool": "autotask_contractRoleCosts_query",
       "purpose": "Query ContractRoleCosts records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_role_costs.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_role_costs.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15022,7 +15254,7 @@
     {
       "operation": "contract_service_adjustments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractServiceAdjustment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractServiceAdjustment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_service_adjustments.getFieldInfo",
       "effect": "read",
@@ -15107,7 +15339,7 @@
     {
       "operation": "contract_service_bundle_adjustments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractServiceBundleAdjustment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractServiceBundleAdjustment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_service_bundle_adjustments.getFieldInfo",
       "effect": "read",
@@ -15208,7 +15440,7 @@
     {
       "operation": "contract_service_bundle_units.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractServiceBundleUnit field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractServiceBundleUnit field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_service_bundle_units.getFieldInfo",
       "effect": "read",
@@ -15241,7 +15473,7 @@
       "operation": "contract_service_bundle_units.query",
       "tool": "autotask_contractServiceBundleUnit_query",
       "purpose": "Query ContractServiceBundleUnit records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_service_bundle_units.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_service_bundle_units.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15344,7 +15576,7 @@
     {
       "operation": "contract_service_bundles.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractServiceBundle field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractServiceBundle field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_service_bundles.getFieldInfo",
       "effect": "read",
@@ -15377,7 +15609,7 @@
       "operation": "contract_service_bundles.query",
       "tool": "autotask_contractServiceBundle_query",
       "purpose": "Query ContractServiceBundle records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_service_bundles.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_service_bundles.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15481,7 +15713,7 @@
     {
       "operation": "contract_service_units.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractServiceUnit field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractServiceUnit field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_service_units.getFieldInfo",
       "effect": "read",
@@ -15514,7 +15746,7 @@
       "operation": "contract_service_units.query",
       "tool": "autotask_contractServiceUnit_query",
       "purpose": "Query ContractServiceUnit records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_service_units.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_service_units.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15636,7 +15868,7 @@
     {
       "operation": "contract_services.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractService field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractService field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_services.getFieldInfo",
       "effect": "read",
@@ -15669,7 +15901,7 @@
       "operation": "contract_services.query",
       "tool": "autotask_contractService_query",
       "purpose": "Query ContractService records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_services.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_services.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15792,7 +16024,7 @@
     {
       "operation": "contract_ticket_purchases.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ContractTicketPurchases field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ContractTicketPurchases field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contract_ticket_purchases.getFieldInfo",
       "effect": "read",
@@ -15825,7 +16057,7 @@
       "operation": "contract_ticket_purchases.query",
       "tool": "autotask_contractTicketPurchase_query",
       "purpose": "Query ContractTicketPurchases records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contract_ticket_purchases.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contract_ticket_purchases.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -15964,7 +16196,7 @@
     {
       "operation": "contracts.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Contract field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Contract field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "contracts.getFieldInfo",
       "effect": "read",
@@ -15997,7 +16229,7 @@
       "operation": "contracts.query",
       "tool": "autotask_contract_query",
       "purpose": "Query Contract records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via contracts.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "contracts.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -16080,6 +16312,25 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "invoices.customRange",
+      "tool": "autotask_invoice_customRange",
+      "purpose": "Get invoices for an explicit invoiceDateTime range.",
+      "preferredWhen": null,
+      "backingOperation": "invoices.customRange",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "from",
+        "to"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "invoices.get",
       "tool": "autotask_invoice_get",
@@ -16117,7 +16368,7 @@
     {
       "operation": "invoices.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Invoice field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Invoice field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "invoices.getFieldInfo",
       "effect": "read",
@@ -16146,6 +16397,150 @@
       "reachReason": "no dedicated tool — the op is still discoverable and invocable by exact name",
       "boundedAlternative": null
     },
+    {
+      "operation": "invoices.last14",
+      "tool": "autotask_invoice_last14",
+      "purpose": "Get invoices for the last 14 days (rolling, invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.last14",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.last24h",
+      "tool": "autotask_invoice_last24h",
+      "purpose": "Get invoices for the last 24 hours (rolling, invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.last24h",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.last30",
+      "tool": "autotask_invoice_last30",
+      "purpose": "Get invoices for the last 30 days (rolling, invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.last30",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.last7",
+      "tool": "autotask_invoice_last7",
+      "purpose": "Get invoices for the last 7 days (rolling, invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.last7",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.last90",
+      "tool": "autotask_invoice_last90",
+      "purpose": "Get invoices for the last 90 days (rolling, invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.last90",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.lastFullFortnight",
+      "tool": "autotask_invoice_lastFullFortnight",
+      "purpose": "Get invoices for the last complete fortnight (invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.lastFullFortnight",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.lastFullMonth",
+      "tool": "autotask_invoice_lastFullMonth",
+      "purpose": "Get invoices for the last complete calendar month (invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.lastFullMonth",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.lastFullQuarter",
+      "tool": "autotask_invoice_lastFullQuarter",
+      "purpose": "Get invoices for the last complete calendar quarter (invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.lastFullQuarter",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "invoices.lastFullWeek",
+      "tool": "autotask_invoice_lastFullWeek",
+      "purpose": "Get invoices for the last complete ISO week (invoiceDateTime).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.lastFullWeek",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "invoices.markupHtml",
       "tool": "autotask_invoice_markup_html",
@@ -16204,7 +16599,7 @@
       "operation": "invoices.query",
       "tool": "autotask_invoice_query",
       "purpose": "Query Invoice records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via invoices.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "invoices.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -16234,6 +16629,22 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "invoices.today",
+      "tool": "autotask_invoice_today",
+      "purpose": "Get invoices for today (invoiceDateTime = now, server date).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.today",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "invoices.update",
       "tool": "autotask_invoice_update",
@@ -16253,6 +16664,22 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "invoices.yesterday",
+      "tool": "autotask_invoice_yesterday",
+      "purpose": "Get invoices for yesterday (invoiceDateTime = previous server day).",
+      "preferredWhen": null,
+      "backingOperation": "invoices.yesterday",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "opportunities.count",
       "tool": "autotask_opportunity_count",
@@ -16342,7 +16769,7 @@
     {
       "operation": "opportunities.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Opportunity field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Opportunity field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "opportunities.getFieldInfo",
       "effect": "read",
@@ -16375,7 +16802,7 @@
       "operation": "opportunities.query",
       "tool": "autotask_opportunity_query",
       "purpose": "Query Opportunity records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via opportunities.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "opportunities.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -16535,7 +16962,7 @@
     {
       "operation": "opportunity_attachments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the OpportunityAttachment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the OpportunityAttachment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "opportunity_attachments.getFieldInfo",
       "effect": "read",
@@ -16568,7 +16995,7 @@
       "operation": "opportunity_attachments.query",
       "tool": "autotask_opportunityAttachment_query",
       "purpose": "Query OpportunityAttachment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via opportunity_attachments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "opportunity_attachments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -16671,7 +17098,7 @@
     {
       "operation": "phases.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Phase field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Phase field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "phases.getFieldInfo",
       "effect": "read",
@@ -16704,7 +17131,7 @@
       "operation": "phases.query",
       "tool": "autotask_phase_query",
       "purpose": "Query Phase records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via phases.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "phases.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -16865,7 +17292,7 @@
     {
       "operation": "project_charges.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ProjectCharge field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ProjectCharge field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "project_charges.getFieldInfo",
       "effect": "read",
@@ -16898,7 +17325,7 @@
       "operation": "project_charges.query",
       "tool": "autotask_projectCharge_query",
       "purpose": "Query ProjectCharge records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via project_charges.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "project_charges.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17039,7 +17466,7 @@
     {
       "operation": "project_notes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the ProjectNote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the ProjectNote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "project_notes.getFieldInfo",
       "effect": "read",
@@ -17072,7 +17499,7 @@
       "operation": "project_notes.query",
       "tool": "autotask_projectNote_query",
       "purpose": "Query ProjectNote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via project_notes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "project_notes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17211,7 +17638,7 @@
     {
       "operation": "projects.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Project field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Project field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "projects.getFieldInfo",
       "effect": "read",
@@ -17244,7 +17671,7 @@
       "operation": "projects.query",
       "tool": "autotask_project_query",
       "purpose": "Query Project records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via projects.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "projects.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17385,7 +17812,7 @@
     {
       "operation": "quote_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the QuoteItem field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the QuoteItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "quote_items.getFieldInfo",
       "effect": "read",
@@ -17418,7 +17845,7 @@
       "operation": "quote_items.query",
       "tool": "autotask_quoteItem_query",
       "purpose": "Query QuoteItem records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via quote_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "quote_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17557,7 +17984,7 @@
     {
       "operation": "quote_locations.getFieldInfo",
       "tool": null,
-      "purpose": "Get the QuoteLocation field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the QuoteLocation field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "quote_locations.getFieldInfo",
       "effect": "read",
@@ -17590,7 +18017,7 @@
       "operation": "quote_locations.query",
       "tool": "autotask_quoteLocation_query",
       "purpose": "Query QuoteLocation records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via quote_locations.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "quote_locations.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17710,7 +18137,7 @@
     {
       "operation": "quotes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Quote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Quote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "quotes.getFieldInfo",
       "effect": "read",
@@ -17743,7 +18170,7 @@
       "operation": "quotes.query",
       "tool": "autotask_quote_query",
       "purpose": "Query Quote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via quotes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "quotes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -17899,7 +18326,7 @@
     {
       "operation": "subscriptions.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Subscription field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Subscription field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "subscriptions.getFieldInfo",
       "effect": "read",
@@ -17932,7 +18359,7 @@
       "operation": "subscriptions.query",
       "tool": "autotask_subscription_query",
       "purpose": "Query Subscription records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via subscriptions.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "subscriptions.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18073,7 +18500,7 @@
     {
       "operation": "task_secondary_resources.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TaskSecondaryResource field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TaskSecondaryResource field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "task_secondary_resources.getFieldInfo",
       "effect": "read",
@@ -18106,7 +18533,7 @@
       "operation": "task_secondary_resources.query",
       "tool": "autotask_taskSecondaryResources_query",
       "purpose": "Query TaskSecondaryResource records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via task_secondary_resources.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "task_secondary_resources.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18209,7 +18636,7 @@
     {
       "operation": "tasks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Task field catalog (types, picklists, required, maxLength) — task/project-module work item.",
+      "purpose": "Get the Task field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog); task/project-module work item.",
       "preferredWhen": null,
       "backingOperation": "tasks.getFieldInfo",
       "effect": "read",
@@ -18242,7 +18669,7 @@
       "operation": "tasks.query",
       "tool": "autotask_task_query",
       "purpose": "Query Task work items (task/project module) with a JSON filter. Not checklist steps; narrow to a project with the projectID filter condition.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via tasks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "tasks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18421,7 +18848,7 @@
     {
       "operation": "ticket_additional_configuration_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketAdditionalConfigurationItem field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketAdditionalConfigurationItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_additional_configuration_items.getFieldInfo",
       "effect": "read",
@@ -18454,7 +18881,7 @@
       "operation": "ticket_additional_configuration_items.query",
       "tool": "autotask_ticketAdditionalConfigurationItem_query",
       "purpose": "Query TicketAdditionalConfigurationItem records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_additional_configuration_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_additional_configuration_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18595,7 +19022,7 @@
     {
       "operation": "ticket_additional_contacts.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketAdditionalContact field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketAdditionalContact field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_additional_contacts.getFieldInfo",
       "effect": "read",
@@ -18628,7 +19055,7 @@
       "operation": "ticket_additional_contacts.query",
       "tool": "autotask_ticketAdditionalContact_query",
       "purpose": "Query TicketAdditionalContact records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_additional_contacts.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_additional_contacts.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18769,7 +19196,7 @@
     {
       "operation": "ticket_attachments.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketAttachment field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketAttachment field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_attachments.getFieldInfo",
       "effect": "read",
@@ -18802,7 +19229,7 @@
       "operation": "ticket_attachments.query",
       "tool": "autotask_ticketAttachment_query",
       "purpose": "Query TicketAttachment records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_attachments.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_attachments.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -18924,7 +19351,7 @@
     {
       "operation": "ticket_change_request_approvals.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketChangeRequestApproval field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketChangeRequestApproval field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_change_request_approvals.getFieldInfo",
       "effect": "read",
@@ -18957,7 +19384,7 @@
       "operation": "ticket_change_request_approvals.query",
       "tool": "autotask_ticketChangeRequestApprovals_query",
       "purpose": "Query TicketChangeRequestApproval records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_change_request_approvals.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_change_request_approvals.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19098,7 +19525,7 @@
     {
       "operation": "ticket_charges.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketCharge field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketCharge field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_charges.getFieldInfo",
       "effect": "read",
@@ -19131,7 +19558,7 @@
       "operation": "ticket_charges.query",
       "tool": "autotask_ticketCharge_query",
       "purpose": "Query TicketCharge records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_charges.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_charges.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19291,7 +19718,7 @@
     {
       "operation": "ticket_checklist_items.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketChecklistItem field catalog (types, picklists, required, maxLength) — ticket checklist step.",
+      "purpose": "Get the TicketChecklistItem field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog); ticket checklist step.",
       "preferredWhen": null,
       "backingOperation": "ticket_checklist_items.getFieldInfo",
       "effect": "read",
@@ -19324,7 +19751,7 @@
       "operation": "ticket_checklist_items.query",
       "tool": "autotask_ticketChecklistItem_query",
       "purpose": "Query TicketChecklistItem records (steps of specific ticket checklists) with a JSON filter. Not Tasks; narrow to a ticket with the ticketID filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_checklist_items.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_checklist_items.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19431,7 +19858,7 @@
     {
       "operation": "ticket_checklist_libraries.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketChecklistLibrary field catalog (types, picklists, required, maxLength) — checklist-library/ticket link.",
+      "purpose": "Get the TicketChecklistLibrary field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog); checklist-library/ticket link.",
       "preferredWhen": null,
       "backingOperation": "ticket_checklist_libraries.getFieldInfo",
       "effect": "read",
@@ -19551,7 +19978,7 @@
     {
       "operation": "ticket_notes.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketNote field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketNote field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_notes.getFieldInfo",
       "effect": "read",
@@ -19584,7 +20011,7 @@
       "operation": "ticket_notes.query",
       "tool": "autotask_ticketNote_query",
       "purpose": "Query TicketNote records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_notes.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_notes.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19726,7 +20153,7 @@
     {
       "operation": "ticket_secondary_resources.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketSecondaryResource field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketSecondaryResource field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_secondary_resources.getFieldInfo",
       "effect": "read",
@@ -19759,7 +20186,7 @@
       "operation": "ticket_secondary_resources.query",
       "tool": "autotask_ticketSecondaryResources_query",
       "purpose": "Query TicketSecondaryResource records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_secondary_resources.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_secondary_resources.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19878,7 +20305,7 @@
     {
       "operation": "ticket_webhooks.getFieldInfo",
       "tool": null,
-      "purpose": "Get the TicketWebhook field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the TicketWebhook field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "ticket_webhooks.getFieldInfo",
       "effect": "read",
@@ -19911,7 +20338,7 @@
       "operation": "ticket_webhooks.query",
       "tool": "autotask_ticketWebhook_query",
       "purpose": "Query TicketWebhook records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via ticket_webhooks.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "ticket_webhooks.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -19981,7 +20408,7 @@
     {
       "operation": "tickets.changeStatus",
       "tool": "autotask_ticket_change_status",
-      "purpose": "Move a ticket to a target status in ONE op — the node carries the ww29 status picklist + the live-discovered per-status required-field map and fills the proven-safe defaults itself (the vendor 500s when a target status has unsatisfied field requirements; the MCP LLM has no way to know the map).",
+      "purpose": "Move a ticket to a target status in ONE op — the node validates targetStatus against the tenant's LIVE status picklist (entity metadata; vendor rejection is the fallback when the metadata is unavailable) + the live-discovered per-status required-field map and fills the proven-safe defaults itself (the vendor 500s when a target status has unsatisfied field requirements; the MCP LLM has no way to know the map).",
       "preferredWhen": "The one-call 'move/close a ticket' for agents: the raw describe + get + update (±500-guess-retry) choreography (4–6 calls today) collapses to this (1 call; 3 wire typical). tickets.update stays the low-level escape hatch.",
       "backingOperation": "tickets.changeStatus",
       "effect": "write",
@@ -20122,7 +20549,7 @@
     {
       "operation": "tickets.getFieldInfo",
       "tool": null,
-      "purpose": "Get the Ticket field catalog (types, picklists, required, maxLength).",
+      "purpose": "Get the Ticket field catalog (types, picklists, required, maxLength) — the full raw field dump (the largest read of the info trio; compact discovery: operations.fieldCatalog).",
       "preferredWhen": null,
       "backingOperation": "tickets.getFieldInfo",
       "effect": "read",
@@ -20173,7 +20600,7 @@
       "operation": "tickets.query",
       "tool": "autotask_ticket_query",
       "purpose": "Query Ticket records with a JSON filter.",
-      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via tickets.getFieldInfo).",
+      "preferredWhen": "Default for filtered reads (the bounded query op). queryAll is the unbounded raw walk — only when you need the ENTIRE set and accept a large payload; default to the bounded query op. Prefer an includeFields subset over the full model (field names via operations.fieldCatalog). latest: N is the latest-N rows (id-descending) when the newest rows are the goal (the vendor has no sort — count-anchored, docs/API.md §4). Prefer a fieldsPreset for common narrow reads (v1: \"billing_join\" on billing_items.query) over a hand-listed includeFields; noteChars: N caps the untrusted-class free-text fields of the resource (notes/descriptions/UDF prose) to N chars each with the explicit … [+N chars truncated] marker (opt-in; the raw result when absent).",
       "backingOperation": "tickets.query",
       "effect": "read",
       "dryRunFirst": false,
@@ -20276,6 +20703,25 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "operations.aggregate",
+      "tool": "autotask_aggregate",
+      "purpose": "Fold any query surface into groups/sums/totals in one bounded call — or count it with the vendor count endpoint when no folding is needed — with blind-surface and truncation honesty",
+      "preferredWhen": "For existence checks and totals over any queryable table (count mode is one call vs a 50k-char page); group-by/sum folds the vendor has no endpoint for; a targeted read is cheaper when you need the rows",
+      "backingOperation": "operations.aggregate",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "operation",
+        "filter"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "operations.blindSurfaces",
       "tool": "autotask_blind_surfaces",
@@ -20308,11 +20754,47 @@
       "reachReason": null,
       "boundedAlternative": null
     },
+    {
+      "operation": "operations.expandResourceNames",
+      "tool": "autotask_expand_resource_names",
+      "purpose": "Resolve resource ids to display names in one opt-in call: per-id resources.get reads (never an id-in query) with blind-marked rows for ids the API user cannot resolve and an opt-in in-memory name cache (each distinct id fetched at most once per cache lifetime).",
+      "preferredWhen": "For name-joining resource ids that other tables expose without names (time_entries / billing_items / tickets resourceID) — fetchByIds for the full records; the bounded query op for filtered resource sets (NOT for lookups on a table that may be blind to the API user — that is exactly when per-id reads are required).",
+      "backingOperation": "operations.expandResourceNames",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "ids"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
+    {
+      "operation": "operations.fetchByIds",
+      "tool": "autotask_fetch_by_ids",
+      "purpose": "Fetch up to 500 single records by id across mixed resources in one call (query-first chunked id-in reads with a per-id get fallback), with per-item typed results.",
+      "preferredWhen": "The default for 'fetch these N known ids in one shot' when N is 20..500 (mixed resources OK — fetchMany is the <=20 single-record GET batch). Wire cost: per resource ceil(uniqueIds/250) query calls + one get per query-missed id — a clean 500-id single-resource batch is 2 wire; a blind-surface worst case is 2 + up to 500 gets (fetchMany parity, never worse). For string identifiers (names, emails, serials) use autotask_resolve_any; for a filtered set use the resource's bounded query op.",
+      "backingOperation": "operations.fetchByIds",
+      "effect": "read",
+      "dryRunFirst": false,
+      "confirm": false,
+      "requiredArgs": [
+        "items"
+      ],
+      "tier": "extended",
+      "reachable": true,
+      "reach": "dedicated tool",
+      "reachReason": null,
+      "boundedAlternative": null
+    },
     {
       "operation": "operations.fetchMany",
       "tool": "autotask_fetch_many",
       "purpose": "Fetch up to 20 single records across mixed resources in one call (tickets + projects + companies in a single batch), with per-item typed results.",
-      "preferredWhen": "The default for 'fetch these N known ids in one shot' (mixed resources OK). For MORE than 20 records or filtered sets use the resource's bounded query op. For string identifiers (names, emails, serials) use autotask_resolve_any.",
+      "preferredWhen": "The default for 'fetch these N known ids in one shot' (mixed resources OK). For MORE than 20 records use operations.fetchByIds (up to 500 ids, mixed resources) or the resource's bounded query op (filtered sets). For string identifiers (names, emails, serials) use autotask_resolve_any.",
       "backingOperation": "operations.fetchMany",
       "effect": "read",
       "dryRunFirst": false,
@@ -20329,8 +20811,8 @@
     {
       "operation": "operations.fieldCatalog",
       "tool": "autotask_field_catalog",
-      "purpose": "Fetch the LIVE vendor field catalog for the named resources in one bounded read call: which fields exist, which are required, types, maxLength, read-only, queryable, picklist-ness — the on-demand refresh for the point-in-time pins (operations.createGuidance, the zoneFacts) when the vendor has moved since they were verified. The wire cost (resources × include) is refused pre-request above the cap, never truncated; with the opt-in fieldCache configured, repeat reads inside the TTL cost no wire. It reports field SHAPE — a picklist VALUE the vendor does not expose to this API user is not resolvable from this catalog (where the vendor does expose values, they are served).",
-      "preferredWhen": "After a write failed with a field-ladder 500, or when the pinned guidance (operations.createGuidance, the zoneFacts pins) predates the vendor's last drift — 'what does the vendor require RIGHT NOW?'. The pinned catalog is the cheap first read and this is the refresh channel (they do not supersede each other). With the opt-in fieldCache configured, repeat reads inside the TTL are zero-wire.",
+      "purpose": "Fetch the LIVE vendor field catalog for the named resources in one bounded read call: which fields exist, which are required, types, maxLength, read-only, queryable, picklist-ness — the on-demand refresh for the point-in-time pins (operations.createGuidance, the zoneFacts) when the vendor has moved since they were verified. The wire cost (resources × include) is refused pre-request above the cap, never truncated; with the opt-in fieldCache configured, repeat reads inside the TTL cost no wire. It reports picklist-ness as a flag (isPickList) and picklist field NAMES (picklistFields); with the opt-in labels: true the projected picklist rows carry the vendor's exposed picklist VALUES (labels next to each value), and without it the op serves no VALUES. A picklist VALUE the vendor does not expose to this API user is not resolvable from this catalog.",
+      "preferredWhen": "After a write failed with a field-ladder 500, or when the pinned guidance (operations.createGuidance, the zoneFacts pins) predates the vendor's last drift — 'what does the vendor require RIGHT NOW?'. The pinned catalog is the cheap first read and this is the refresh channel (they do not supersede each other). With the opt-in fieldCache configured, repeat reads inside the TTL are zero-wire. Field discovery for any resource: names/types/required/readonly/queryable/picklist-ness in one bounded call — prefer it over per-resource getFieldInfo.",
       "backingOperation": "operations.fieldCatalog",
       "effect": "read",
       "dryRunFirst": false,
@@ -20396,10 +20878,10 @@
     }
   ],
   "counts": {
-    "registryOperations": 1183,
+    "registryOperations": 1212,
     "coreOperations": 10,
-    "catalogRows": 1173,
-    "withTool": 810,
+    "catalogRows": 1202,
+    "withTool": 839,
     "readModeOnly": 363,
     "neverExposed": 0,
     "metaTools": 5