deepline 0.3.105 → 0.3.107

package/dist/bundling-sources/shared_libs/observability/scheduled-jobs.json

@@ -9,7 +9,13 @@
     "locator": "schedule:",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Compare Convex, Metronome, Snowflake, and Stripe billing truth and retain a private evidence report.",
+    "successDefinition": "All required sources are queried and a complete parity report is written; zero mismatches is healthy.",
+    "failureDefinition": "A source/report/notification fails or the audit contains an unresolved billing mismatch.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.billing-parity-watchdog",
@@ -21,7 +27,13 @@
     "locator": "schedule:",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Verify that the nightly billing parity audit produced today's complete report.",
+    "successDefinition": "The expected-date report exists, is complete, and passes integrity checks.",
+    "failureDefinition": "The report is missing, incomplete, stale, or invalid.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.ci-health-analytics",
@@ -33,7 +45,13 @@
     "locator": "schedule:",
     "owner": "data-platform",
     "criticality": "reporting",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Build validated CI health snapshots, DuckDB marts, and Ojos serving artifacts.",
+    "successDefinition": "Snapshots are complete, marts validate, and the versioned serving pointer advances atomically.",
+    "failureDefinition": "A snapshot, dbt/mart validation, artifact, or pointer publish fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.cleanup-stale-preview-infrastructure",
@@ -45,7 +63,13 @@
     "locator": "schedule:",
     "owner": "platform",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Reclaim closed or abandoned Preview Convex, Inngest, Fly, and Neon infrastructure without touching active leases.",
+    "successDefinition": "Inventory and guarded cleanup finish; zero eligible stale resources is a valid no-op.",
+    "failureDefinition": "Credentials/inventory are unavailable, a lease is ambiguous, or cleanup/delete fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.convex-data-sync-to-snowflake",
@@ -57,7 +81,13 @@
     "locator": "schedule:",
     "owner": "data-platform",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Copy the Convex mirror incrementally into Snowflake and advance the mirror continuation.",
+    "successDefinition": "Every selected COPY/continuation completes and the Snowflake watermark advances to the source head.",
+    "failureDefinition": "A source page, COPY, continuation, or watermark update fails or leaves the mirror behind.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.daily-slack-report",
@@ -69,7 +99,13 @@
     "locator": "schedule:",
     "owner": "product",
     "criticality": "reporting",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Read the canonical Snowflake mirror and post the daily usage report to the reporting Slack channel.",
+    "successDefinition": "The query returns a coherent summary and Slack accepts it; an empty report is explicitly rendered as no activity.",
+    "failureDefinition": "Snowflake cannot provide the summary or Slack rejects/loses the report.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.operational-alerts",
@@ -81,7 +117,13 @@
     "locator": "schedule:",
     "owner": "reliability",
     "criticality": "critical",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Reconcile Axiom and Better Stack alert definitions and synchronize production thresholds.",
+    "successDefinition": "Both providers match the checked-in rules and threshold changes are verified.",
+    "failureDefinition": "Credentials/provider writes fail or read-back disagrees with the desired alert set.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.eval-agent-cli-matrix",
@@ -93,7 +135,13 @@
     "locator": "schedule:",
     "owner": "evals",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Run the scheduled agent/CLI evaluation matrix and publish its consolidated evidence viewer.",
+    "successDefinition": "All required lanes finish, raw results are complete, and the consolidated report is published.",
+    "failureDefinition": "A lane, credential, result merge, viewer publish, or Slack consolidation fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.fly-pr-preview-cleanup",
@@ -105,7 +153,13 @@
     "locator": "schedule:",
     "owner": "platform",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Reconcile completed PR Preview Machines and release capacity only after the per-app lease fence allows it.",
+    "successDefinition": "Fly inventory is reconciled and eligible terminal/closed PR capacity is released; no eligible app is a valid no-op.",
+    "failureDefinition": "Inventory, lease resolution, scale-down, or empty-app cleanup fails or is ambiguous.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.log-triage",
@@ -117,7 +171,13 @@
     "locator": "schedule:",
     "owner": "reliability",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Run a fresh-context log triage pass and publish its report and context bundle for on-call follow-up.",
+    "successDefinition": "The triage agent completes, durable artifacts are uploaded, and the internal report is posted.",
+    "failureDefinition": "Agent execution, log access, artifact upload, or report publication fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.metronome-catalog-ci-cleanup",
@@ -129,7 +189,13 @@
     "locator": "schedule:",
     "owner": "billing",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Refund disposable Metronome sandbox payments and archive CI customers while retaining reusable catalog fixtures.",
+    "successDefinition": "Every targeted resource is handled, or an empty inventory is recorded as no-op.",
+    "failureDefinition": "A refund/archive is incomplete, a target is ambiguous, or cleanup evidence is not retained.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.metronome-catalog-integrity",
@@ -141,7 +207,13 @@
     "locator": "schedule:",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Plan, provision, read back, and exercise the exact Metronome catalog revision used by CI and releases.",
+    "successDefinition": "The approved fingerprint converges in the provider, read-back evidence is retained, and the disposable flow passes.",
+    "failureDefinition": "The revision is unapproved, read-back diverges, or acquisition/usage/renewal proof fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.neon-ingestion-cleanup",
@@ -153,7 +225,13 @@
     "locator": "schedule:",
     "owner": "data-platform",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Inventory and reclaim stale non-production Neon ingestion projects while honoring age, lease, and exact-target fences.",
+    "successDefinition": "The requested inventory/reclaim completes and the free-capacity target is met; no eligible project is valid no-op.",
+    "failureDefinition": "Credentials/inventory are incomplete, a lease blocks safe action, or deletion/receipt fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.provider-dashboard",
@@ -165,7 +243,13 @@
     "locator": "schedule:",
     "owner": "integrations",
     "criticality": "reporting",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Build nightly provider/integration health dashboard artifacts from the current catalog and analytics sources.",
+    "successDefinition": "Catalog build and provider pipeline finish and publish a complete report artifact.",
+    "failureDefinition": "Catalog generation, source queries, dashboard computation, or publication fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.refresh-openapi-snapshots",
@@ -177,7 +261,13 @@
     "locator": "schedule:",
     "owner": "integrations",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Refresh live OpenAPI snapshots and open or update the generated refresh pull request.",
+    "successDefinition": "Snapshots/artifacts are current and the refresh PR is updated; no diff is a valid no-op.",
+    "failureDefinition": "Refresh, generation, validation, or PR create/update fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.runtime-lifecycle",
@@ -189,7 +279,13 @@
     "locator": "schedule:",
     "owner": "runtime",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Exercise the canonical Daytona/Play runtime lifecycle against the exact healthy Preview target and retain load evidence.",
+    "successDefinition": "Target preparation, lifecycle bursts, correctness/latency checks, cleanup, and reporting complete.",
+    "failureDefinition": "Target resolution, lifecycle execution, evaluation, cleanup, or reporting fails.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.runtime-shape",
@@ -205,7 +301,13 @@
     "locator": "schedule:",
     "owner": "runtime",
     "criticality": "operational",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Run the exact-SHA Play runtime dataset-shape matrix across isolated Preview lanes.",
+    "successDefinition": "All required lanes pass and every lane is cleaned up with detailed results retained.",
+    "failureDefinition": "A lane cannot prepare, violates a shape invariant, or cleanup/evidence is incomplete.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "github.refresh-vitest-shard-weights",
@@ -217,7 +319,13 @@
     "locator": "schedule:",
     "owner": "platform",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Measure every Vitest shard, rebuild timing weights, and open or update the refresh PR.",
+    "successDefinition": "All shard timings are present, predicted balance is valid, and the PR is updated; no material change is a valid no-op.",
+    "failureDefinition": "A shard fails/is missing, the manifest is incomplete, or the PR cannot be updated.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.charge-reconcile",
@@ -229,7 +337,13 @@
     "locator": "/api/v2/cron/charge-reconcile",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Reconcile pending charge reservations, usage truth, provider billing events, and bounded continuation backlog.",
+    "successDefinition": "Due financial work settles or is durably parked with recovery, witnesses are recorded, and backlog advances.",
+    "failureDefinition": "A ledger/provider mismatch is unresolved, parked work lacks recovery, or continuation cannot advance.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.metronome-usage-export",
@@ -241,7 +355,13 @@
     "locator": "/api/v2/cron/metronome-usage-export",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Drain the commercial usage export outbox to Metronome and evaluate export freshness.",
+    "successDefinition": "Due usage is delivered or the outbox is empty, and age/error health is within budget.",
+    "failureDefinition": "Delivery fails, export stays stale beyond budget, or health cannot persist.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-health",
@@ -253,7 +373,13 @@
     "locator": "/api/v2/cron/provider-health",
     "owner": "integrations",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Probe every configured provider, persist the stabilized health snapshot, and emit transition/burn-rate signals.",
+    "successDefinition": "All probes complete, snapshot/history persist, and degraded providers are explicitly classified.",
+    "failureDefinition": "Probes are incomplete, persistence fails, or a transition notification cannot reconcile.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-health-spend-digest",
@@ -265,7 +391,13 @@
     "locator": "/api/v2/cron/provider-health-spend-digest",
     "owner": "integrations",
     "criticality": "reporting",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Turn latest provider health and monthly usage into the scheduled spend digest.",
+    "successDefinition": "A current snapshot is summarized and delivered; no providers is explicitly reported as no data.",
+    "failureDefinition": "Sources are missing/stale, digest construction fails, or delivery is rejected.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.daily-activity-summary",
@@ -277,7 +409,13 @@
     "locator": "/api/v2/cron/daily-activity-summary",
     "owner": "product",
     "criticality": "reporting",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Summarize the previous reporting window from Convex usage and post the daily activity event to Slack.",
+    "successDefinition": "The correct timezone window is queried and one coherent summary is accepted by Slack.",
+    "failureDefinition": "The usage query or summary post fails, producing no trustworthy report.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.sdk-cli-activity-digest",
@@ -289,7 +427,13 @@
     "locator": "/api/v2/cron/sdk-cli-activity-digest",
     "owner": "reliability",
     "criticality": "reporting",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Aggregate SDK and CLI telemetry for the requested bucket and publish the activity digest.",
+    "successDefinition": "The bucket is finalized and delivered; an empty bucket is explicitly recorded as no activity.",
+    "failureDefinition": "Telemetry aggregation, finalization, or digest delivery fails.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.billing-integrity",
@@ -301,7 +445,13 @@
     "locator": "/api/v2/cron/billing-integrity",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Check usage-rollup freshness and reconcile sampled organization ledger totals against activity.",
+    "successDefinition": "Sampled organizations reconcile within tolerance and rollup lag stays below budget.",
+    "failureDefinition": "Lag breaches budget, an organization has an unresolved mismatch, or evidence is incomplete.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.billing-autobill",
@@ -313,7 +463,13 @@
     "locator": "/api/v2/cron/billing-autobill",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Evaluate enabled autobill settings and charge each due organization with durable outcome recording.",
+    "successDefinition": "Every selected setting is evaluated and persisted; no due organizations/all-skipped is a valid no-op.",
+    "failureDefinition": "A charge runner errors or an evaluation/outcome cannot be persisted.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.signal-radar-dead-letter-digest",
@@ -325,7 +481,13 @@
     "locator": "/api/v2/cron/signal-radar-dead-letter-digest",
     "owner": "monitors",
     "criticality": "operational",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Scan monitored organizations for signal-radar ingestion dead letters, alert on backlog, and advance recovery cursors.",
+    "successDefinition": "All selected organizations are scanned, backlog is reported, and the cursor advances only after complete evidence.",
+    "failureDefinition": "A scan/connection or alert fails, or the cursor advances without complete evidence.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-monitor-audit-sweep",
@@ -337,7 +499,13 @@
     "locator": "/api/v2/cron/provider-monitor-audit-sweep",
     "owner": "monitors",
     "criticality": "operational",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Sweep due provider-monitor audit records and apply the audit reconciliation policy.",
+    "successDefinition": "Every due audit is processed and persisted; zero due audits is a valid no-op.",
+    "failureDefinition": "A due audit cannot be processed or its state/evidence remains unresolved.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-monitor-policy-sweep",
@@ -349,19 +517,13 @@
     "locator": "/api/v2/cron/provider-monitor-policy-sweep",
     "owner": "monitors",
     "criticality": "operational",
-    "failureAlert": "vercel_route_wrapper"
-  },
-  {
-    "id": "vercel.provider-monitor-direct-billing-suspension",
-    "name": "Provider monitor direct billing suspension",
-    "scheduler": "vercel_cron",
-    "cadence": "*/5 * * * *",
-    "timezone": "UTC",
-    "source": "apps/deepline-api/vercel.json",
-    "locator": "/api/v2/cron/provider-monitor-direct-billing-suspension",
-    "owner": "monitors",
-    "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Sweep provider-monitor policy state and converge monitors to declared desired policy.",
+    "successDefinition": "Every due candidate is evaluated and persisted; zero due candidates is a valid no-op.",
+    "failureDefinition": "Policy evaluation/persistence fails or a monitor remains unresolved.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-monitor-batch-operation-sweep",
@@ -373,7 +535,13 @@
     "locator": "/api/v2/cron/provider-monitor-batch-operation-sweep",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Recover and advance provider-monitor batch operations that are queued, delayed, or stuck.",
+    "successDefinition": "Every due operation completes, retries durably, or is terminally classified; no due operations is a valid no-op.",
+    "failureDefinition": "An operation is left stuck without a next action or recovery state cannot be written.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.provider-monitor-webhook-reconciliation",
@@ -385,7 +553,13 @@
     "locator": "/api/v2/cron/provider-monitor-webhook-reconciliation",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Reconcile provider-monitor webhook receipts with durable operation state and continuation bridges.",
+    "successDefinition": "Selected receipts agree with operation state and anomalies have evidence; no receipts is a valid no-op.",
+    "failureDefinition": "Bridge/scan/reconciliation write fails or an unresolved mismatch remains.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.scheduled-work-health-snapshot",
@@ -397,7 +571,13 @@
     "locator": "/api/v2/cron/scheduled-work-health-snapshot",
     "owner": "reliability",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Synchronize scheduled-work catalog definitions used by the operational health surface.",
+    "successDefinition": "Every catalog definition is accepted and the current snapshot is complete.",
+    "failureDefinition": "Validation or synchronization fails, leaving stale/incomplete dashboard semantics.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.provider-monitor-webhook-reconciliation-continuation",
@@ -409,7 +589,13 @@
     "locator": "dispatchProviderMonitorWebhookReconciliationContinuation",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Dispatch one bounded provider-monitor webhook reconciliation continuation owned by Convex.",
+    "successDefinition": "The exact continuation is accepted or already terminal/complete, with durable duplicate protection.",
+    "failureDefinition": "It cannot be claimed/dispatched or durable state is ambiguous.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.provider-monitor-batch-operation-continuation",
@@ -421,7 +607,13 @@
     "locator": "dispatchProviderMonitorBatchOperationContinuation",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Dispatch one bounded provider-monitor batch-operation continuation to the provider executor.",
+    "successDefinition": "The exact continuation is accepted or already terminal and the claim fence is durable.",
+    "failureDefinition": "It cannot be claimed/dispatched or its state transition is ambiguous.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.provider-monitor-global-audit-continuation",
@@ -433,7 +625,13 @@
     "locator": "dispatchProviderMonitorGlobalAuditContinuation",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Advance one leased page of the global provider-monitor audit and schedule the next bounded step.",
+    "successDefinition": "The page is processed and the next step is scheduled or audit is terminally recorded.",
+    "failureDefinition": "Page completion, continuation scheduling, or lease/state persistence fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.provider-monitor-global-audit-daily",
@@ -445,7 +643,13 @@
     "locator": "provider monitor global audit daily",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Claim and run the daily global provider-monitor audit across the shared account inventory.",
+    "successDefinition": "The inventory is traversed and anomalies are classified/persisted; an empty inventory is a valid no-op.",
+    "failureDefinition": "Lease, inventory traversal, provider read, or anomaly persistence fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.provider-monitor-credit-recovery",
@@ -457,7 +661,13 @@
     "locator": "recoverProviderMonitorsAfterCreditBalanceChange",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Recover provider monitors whose billing suspension can be lifted after a positive credit-balance change.",
+    "successDefinition": "Idempotent recovery intent applies to the target organization; no eligible monitors is a valid no-op.",
+    "failureDefinition": "Recovery action fails or idempotency/state transition is not durable.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "inngest.workflow-run-watchdog",
@@ -469,7 +679,13 @@
     "locator": "workflow_run_watchdog_v1",
     "owner": "workflows",
     "criticality": "operational",
-    "failureAlert": "inngest_on_failure"
+    "purpose": "Fail workflow runs beyond their stale-progress budget and emit alertable cron-run events.",
+    "successDefinition": "Stale statuses drain within the bounded budget and each failed run has a terminal event; no stale runs is a valid no-op.",
+    "failureDefinition": "A stale run cannot be terminalized/alerted or the sweep fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "inngest_on_failure",
+    "domainReportRequired": true
   },
   {
     "id": "inngest.workflow-cron-tick",
@@ -481,7 +697,13 @@
     "locator": "workflow_cron_tick_v1",
     "owner": "workflows",
     "criticality": "critical",
-    "failureAlert": "inngest_on_failure"
+    "purpose": "Find due workflow cron bindings and enqueue their next workflow runs.",
+    "successDefinition": "Every due binding is enqueued or durably skipped by its control-plane fence.",
+    "failureDefinition": "Scan/enqueue fails or due bindings lack a durable disposition.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "inngest_on_failure",
+    "domainReportRequired": true
   },
   {
     "id": "inngest.play-cron-tick",
@@ -493,7 +715,13 @@
     "locator": "play_cron_tick_v1",
     "owner": "runtime",
     "criticality": "critical",
-    "failureAlert": "inngest_on_failure"
+    "purpose": "Find due Play schedules and enqueue their next Play executions.",
+    "successDefinition": "Every due Play is enqueued or durably skipped by its schedule fence.",
+    "failureDefinition": "Scan/enqueue fails or a due Play lacks a durable disposition.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "inngest_on_failure",
+    "domainReportRequired": true
   },
   {
     "id": "inngest.ingestion-schedule-tick",
@@ -505,7 +733,13 @@
     "locator": "ingestion_schedule_tick_v1",
     "owner": "data-platform",
     "criticality": "operational",
-    "failureAlert": "inngest_on_failure"
+    "purpose": "Scan active ingestion planes and queue subscriptions whose schedule mirror says work is due.",
+    "successDefinition": "Every eligible plane is evaluated, due subscriptions are queued, and stale/missing mirrors are surfaced.",
+    "failureDefinition": "A plane cannot be read/repaired or due work is lost without a durable disposition.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "inngest_on_failure",
+    "domainReportRequired": true
   },
   {
     "id": "inngest.ingestion-launch-tick",
@@ -517,7 +751,13 @@
     "locator": "ingestion_launch_tick_v1",
     "owner": "data-platform",
     "criticality": "operational",
-    "failureAlert": "inngest_on_failure"
+    "purpose": "Recover stale ingestion launch/execution jobs and dispatch queued jobs.",
+    "successDefinition": "Every eligible queued job launches or is durably left for retry, with stale jobs recovered and diagnosed.",
+    "failureDefinition": "A queued job remains stuck without evidence or scan/dispatch fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "inngest_on_failure",
+    "domainReportRequired": true
   },
   {
     "id": "convex.fold-usage-rollups",
@@ -529,7 +769,13 @@
     "locator": "fold usage rollups",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Fold pending usage-rollup pages into the single-writer aggregate and schedule bounded catch-up.",
+    "successDefinition": "All pages fold or none remain; continuation is scheduled when more work exists.",
+    "failureDefinition": "A page cannot fold or the continuation cursor is lost.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.record-neon-managed-project-usage",
@@ -541,7 +787,31 @@
     "locator": "record Neon managed project usage",
     "owner": "billing",
     "criticality": "operational",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Record current usage for managed Neon projects at daily granularity.",
+    "successDefinition": "Every eligible project is measured and recorded; no eligible projects is a valid no-op.",
+    "failureDefinition": "A project cannot be measured or its usage record fails to persist.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
+  },
+  {
+    "id": "convex.observe-operational-queue-health",
+    "name": "Observe operational queue health",
+    "scheduler": "convex_cron",
+    "cadence": "every 60 seconds",
+    "timezone": "UTC",
+    "source": "packages/convex/crons.ts",
+    "locator": "observe operational queue health",
+    "owner": "reliability",
+    "criticality": "critical",
+    "purpose": "Measure every durable queue through the scheduler-neutral depth, age, flow, and DLQ contract.",
+    "successDefinition": "Every cataloged queue publishes a complete measurement and the observer completes.",
+    "failureDefinition": "The observer is missed/stuck, a queue is silently omitted, or snapshots cannot persist.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.cleanup-expired-mcp-oauth-codes",
@@ -553,7 +823,13 @@
     "locator": "cleanup expired MCP OAuth codes",
     "owner": "auth",
     "criticality": "maintenance",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Delete expired MCP OAuth authorization codes and refresh tokens.",
+    "successDefinition": "All expired credentials are deleted and counted; zero expired credentials is a valid no-op.",
+    "failureDefinition": "The expiry scan/deletion fails, leaving expired credentials retained.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.cleanup-expired-analytics-ip2company-cache",
@@ -565,7 +841,13 @@
     "locator": "cleanup expired Analytics IP2Company cache entries",
     "owner": "monitors",
     "criticality": "maintenance",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Delete expired salted IP-to-company cache entries within the bounded retention window.",
+    "successDefinition": "The expiry scan completes and all eligible entries are deleted; zero eligible entries is a valid no-op.",
+    "failureDefinition": "Cache scan/deletion fails or retention cleanup stops early.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.cleanup-expired-domain-health-cache",
@@ -577,7 +859,13 @@
     "locator": "cleanup expired domain health cache entries",
     "owner": "integrations",
     "criticality": "maintenance",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Delete expired domain-health cache entries in bounded batches.",
+    "successDefinition": "Batches reach completion; zero expired entries is a valid no-op.",
+    "failureDefinition": "A batch deletion fails or cleanup stops before the backlog is represented.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.salesforce-oauth-refresh-maintenance",
@@ -589,7 +877,13 @@
     "locator": "maintain Salesforce OAuth refresh tokens",
     "owner": "integrations",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Refresh Salesforce OAuth tokens before expiry and preserve each connection's rotating refresh-token chain.",
+    "successDefinition": "Every due connection refreshes or is explicitly classified as customer re-auth required, with continuation for remaining pages.",
+    "failureDefinition": "Refresh dispatch fails, token-chain persistence fails, or continuation state is lost.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.recover-notification-delivery-work",
@@ -601,7 +895,13 @@
     "locator": "recover notification delivery work",
     "owner": "platform",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Recover scheduled notification digest buckets and deliveries ready for retry.",
+    "successDefinition": "Every recoverable delivery advances or is durably retried; zero recoverable deliveries is a valid no-op.",
+    "failureDefinition": "A delivery remains stuck without retry/terminal disposition or recovery fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.fail-stale-play-deployments",
@@ -613,7 +913,13 @@
     "locator": "fail stale play deployments",
     "owner": "platform",
     "criticality": "operational",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Terminalize Play deployments past deploy-token lifetime and prune expired webhook dedupe rows.",
+    "successDefinition": "Stale deployments fail and eligible dedupe rows prune; zero stale records is a valid no-op.",
+    "failureDefinition": "A deployment cannot terminalize or pruning fails, leaving state stuck.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "play.deepline-play-execution-heartbeat",
@@ -625,7 +931,13 @@
     "locator": "deepline-play-execution-heartbeat",
     "owner": "runtime",
     "criticality": "critical",
-    "failureAlert": "better_stack_heartbeat"
+    "purpose": "Exercise the Play execution heartbeat that keeps active runs and queue leases alive.",
+    "successDefinition": "Active executions receive heartbeats and idle/terminal runs are reported; no active executions is a valid no-op.",
+    "failureDefinition": "Heartbeat dispatch, lease refresh, or terminal reporting fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "better_stack_heartbeat",
+    "domainReportRequired": true
   },
   {
     "id": "convex.monitor-fleet-convergence-tick",
@@ -637,7 +949,13 @@
     "locator": "monitor fleet convergence tick",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Claim due monitor fleets, dispatch bounded provider work, and request periodic source refresh.",
+    "successDefinition": "Every claim dispatches or durably fails and refresh work is accounted; no due fleets is a valid no-op.",
+    "failureDefinition": "A claim is stranded, provider dispatch lacks recovery, or convergence state cannot persist.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "vercel.play-webhook-receipt-recovery",
@@ -649,7 +967,13 @@
     "locator": "/api/v2/cron/play-webhook-receipt-recovery",
     "owner": "runtime",
     "criticality": "critical",
-    "failureAlert": "vercel_route_wrapper"
+    "purpose": "Recover Play webhook receipts whose admission/dispatch continuation is delayed or stranded.",
+    "successDefinition": "Every eligible receipt is retried or terminally classified with evidence; no eligible receipts is a valid no-op.",
+    "failureDefinition": "A receipt cannot be safely recovered, remains stranded, or evidence cannot persist.",
+    "startGraceMs": 300000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "vercel_route_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.cleanup-expired-notification-audit-records",
@@ -661,7 +985,13 @@
     "locator": "cleanup expired notification audit records",
     "owner": "platform",
     "criticality": "maintenance",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Delete notification audit records past their retention window.",
+    "successDefinition": "All eligible records are deleted; zero eligible records is a valid no-op.",
+    "failureDefinition": "Retention scan/deletion fails, leaving expired notification evidence indefinitely.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "github.modal-ci-runner-weekly-proof",
@@ -673,7 +1003,13 @@
     "locator": "schedule:",
     "owner": "platform",
     "criticality": "maintenance",
-    "failureAlert": "github_workflow_run"
+    "purpose": "Lint, test, deploy, and prove the pinned Modal CI runner image and isolated smoke path.",
+    "successDefinition": "Pinned package lint/tests, control-plane/image deploy, and exact-image credential-isolation smoke all pass.",
+    "failureDefinition": "Any lint/test/deploy/health/smoke/cleanup proof fails or image identity is wrong.",
+    "startGraceMs": 600000,
+    "runTimeoutMs": 3600000,
+    "failureAlert": "github_workflow_run",
+    "domainReportRequired": true
   },
   {
     "id": "convex.async-usage-recovery-continuation",
@@ -685,7 +1021,13 @@
     "locator": "dispatchAsyncUsageRecoveryContinuation",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Continue bounded recovery of asynchronous usage records not admitted or settled on the first pass.",
+    "successDefinition": "The recovery batch processes or durably reschedules until complete; an empty batch is a valid no-op.",
+    "failureDefinition": "A usage record remains unrecovered without a next step or continuation state is lost.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.charge-reconciliation-backlog-continuation",
@@ -697,7 +1039,13 @@
     "locator": "dispatchChargeReconciliationBacklogContinuation",
     "owner": "billing",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Continue bounded reconciliation of charge work beyond the primary cron batch.",
+    "successDefinition": "The backlog page settles, retries durably, or is terminally classified.",
+    "failureDefinition": "Backlog work is stuck, a financial transition is ambiguous, or continuation scheduling fails.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   },
   {
     "id": "convex.monitor-deployment-job-tick",
@@ -709,6 +1057,12 @@
     "locator": "monitor deployment job tick",
     "owner": "monitors",
     "criticality": "critical",
-    "failureAlert": "convex_action_wrapper"
+    "purpose": "Lease and dispatch one legacy scalar monitor deployment job; batch jobs use their continuation runner.",
+    "successDefinition": "The due lease dispatches exactly once or no due job is a valid no-op; lease is released/terminally recorded.",
+    "failureDefinition": "A claimed job cannot dispatch or its lease remains stranded/ambiguous.",
+    "startGraceMs": 120000,
+    "runTimeoutMs": 1800000,
+    "failureAlert": "convex_action_wrapper",
+    "domainReportRequired": true
   }
 ]