@talonic/docs 0.17.0 → 0.17.1

package/dist/index.js

@@ -5052,6 +5052,7 @@ var NAV_SECTIONS2 = [
       { id: "field-registry", label: "Field Registry" },
       { id: "tier-system", label: "Tier System" },
       { id: "semantic-clusters", label: "Semantic Clusters" },
+      { id: "field-resolution", label: "Field Resolution" },
       { id: "master-instructions", label: "Master Instructions" }
     ]
   },
@@ -5115,13 +5116,22 @@ var NAV_SECTIONS2 = [
       { id: "document-graph", label: "Document Graph" }
     ]
   },
+  {
+    id: "data-products",
+    label: "Data Products",
+    children: [
+      { id: "dataset-templates", label: "Dataset Templates" },
+      { id: "assemblies", label: "Assemblies" }
+    ]
+  },
   {
     id: "validation-quality",
     label: "Validation & Quality",
     children: [
       { id: "validation-checks", label: "Validation Checks" },
       { id: "ground-truth", label: "Golden Samples" },
-      { id: "approval-gates", label: "Approval Gates" }
+      { id: "approval-gates", label: "Approval Gates" },
+      { id: "approval-queue", label: "Approval Queue" }
     ]
   },
   {
@@ -5530,6 +5540,25 @@ function PlatformGuide({ LinkComponent }) {
       ] }),
       /* @__PURE__ */ jsx6(SubHeading, { id: "semantic-clusters", children: "Semantic Clusters" }),
       /* @__PURE__ */ jsx6(P, { children: 'Fields with similar meanings are automatically grouped using AI embeddings. For example, "Vendor Name", "Supplier Name", and "Company Name" cluster together. You can manually merge or split clusters from the Field Map view.' }),
+      /* @__PURE__ */ jsx6(SubHeading, { id: "field-resolution", children: "Field Resolution" }),
+      /* @__PURE__ */ jsx6(P, { children: "When a document is processed, each extracted field is resolved against the registry using a three-band matching model. The bands determine whether a match is accepted automatically, flagged for confirmation, or treated as a new field." }),
+      /* @__PURE__ */ jsx6(
+        ParamTable,
+        {
+          params: [
+            { name: "Auto band", type: "\u2265 0.80 similarity", description: "High-confidence match. The field is linked to the existing registry entry and its occurrence count is incremented." },
+            { name: "Confirm band", type: "0.50 \u2013 0.79", description: "Candidate match. The field is linked but flagged for manual review in the reconciliation queue." },
+            { name: "New band", type: "< 0.50", description: "No match found. A new Tier 3 field and cluster are created in the registry." }
+          ]
+        }
+      ),
+      /* @__PURE__ */ jsx6(P, { children: "Resolution runs concurrently across documents. Each document\u2019s fields are resolved in an isolated transaction to prevent lock contention. Occurrence rates are updated after each transaction commits, keeping the registry eventually consistent without blocking concurrent ingestion." }),
+      /* @__PURE__ */ jsxs6(Callout, { children: [
+        "Pending confirmations from the confirm band appear in",
+        " ",
+        /* @__PURE__ */ jsx6("strong", { children: "Resolution \u2192 Pending Confirmations" }),
+        ". Accept to merge into an existing cluster, or reject to create a new field."
+      ] }),
       /* @__PURE__ */ jsx6(SubHeading, { id: "master-instructions", children: "Master Instructions" }),
       /* @__PURE__ */ jsxs6(P, { children: [
         "As the same field is extracted from many documents, AI synthesizes a ",
@@ -5876,6 +5905,23 @@ function PlatformGuide({ LinkComponent }) {
       /* @__PURE__ */ jsx6(P, { children: "Each case detail page shows: documents, shared entities, evidence chain (which field keys produced each connection), timeline, and auto-generated AI narration." }),
       /* @__PURE__ */ jsx6(SubHeading, { id: "document-graph", children: "Document Graph" }),
       /* @__PURE__ */ jsx6(P, { children: "The Document Graph provides a visual D3-force layout of the bipartite graph. Toggle between graph and list views from the Cases page. Case templates are auto-discovered after 3+ cases form \u2014 they identify recurring document type patterns." }),
+      /* @__PURE__ */ jsx6(SectionHeading, { id: "data-products", children: "Data Products" }),
+      /* @__PURE__ */ jsx6(P, { children: "Data Products package your extraction and resolution outputs into reusable, shareable artifacts. Dataset Templates define the shape; Assemblies combine sources into a single structured dataset." }),
+      /* @__PURE__ */ jsx6(SubHeading, { id: "dataset-templates", children: "Dataset Templates" }),
+      /* @__PURE__ */ jsx6(P, { children: "A dataset template defines a reusable output specification: which schema to use, which document types to include, column mappings, and default transforms. Templates ensure consistent data product generation across teams and time periods." }),
+      /* @__PURE__ */ jsxs6(P, { children: [
+        "Navigate to ",
+        /* @__PURE__ */ jsx6("strong", { children: "Data Products \u2192 Dataset Templates" }),
+        " to manage templates. Each template is linked to a user schema and can be versioned independently. When creating a new job, select a template instead of configuring the output from scratch."
+      ] }),
+      /* @__PURE__ */ jsx6(SubHeading, { id: "assemblies", children: "Assemblies" }),
+      /* @__PURE__ */ jsx6(P, { children: "An assembly combines documents from one or more sources into a single structured dataset based on a template. Assemblies track their constituent documents, source counts, and processing status." }),
+      /* @__PURE__ */ jsxs6(P, { children: [
+        "Navigate to ",
+        /* @__PURE__ */ jsx6("strong", { children: "Data Products \u2192 Assemblies" }),
+        " to view and create assemblies. Each assembly shows its document count, linked schema, processing status, and the date it was created."
+      ] }),
+      /* @__PURE__ */ jsx6(Callout, { children: "Assemblies are the recommended way to produce production datasets. They provide a single audit trail from source documents through extraction, resolution, and validation to the final output." }),
       /* @__PURE__ */ jsx6(SectionHeading, { id: "validation-quality", children: "Validation & Quality" }),
       /* @__PURE__ */ jsx6(P, { children: "Multiple layers of quality assurance: automated validation checks, ground truth benchmarking, and configurable approval gates." }),
       /* @__PURE__ */ jsx6(SubHeading, { id: "validation-checks", children: "Validation Checks" }),
@@ -5895,6 +5941,13 @@ function PlatformGuide({ LinkComponent }) {
         /* @__PURE__ */ jsx6(InlineCode, { children: "result.approved" }),
         " signal to a destination to only ship approved rows to your downstream systems."
       ] }),
+      /* @__PURE__ */ jsx6(SubHeading, { id: "approval-queue", children: "Approval Queue" }),
+      /* @__PURE__ */ jsxs6(P, { children: [
+        "Results that do not meet auto-approval thresholds are routed to the Approval Queue. Navigate to ",
+        /* @__PURE__ */ jsx6("strong", { children: "Review \u2192 Approval Queue" }),
+        " to see all pending items. Each row shows the source document, schema, confidence score, and whether the result has been flagged by a validation rule."
+      ] }),
+      /* @__PURE__ */ jsx6(P, { children: "Filter the queue by status (pending, flagged), schema, or confidence range. Click \u201CReview\u201D on any row to inspect the extracted values, provenance trails, and validation check results before approving or rejecting." }),
       /* @__PURE__ */ jsx6(SectionHeading, { id: "delivery", children: "Delivery" }),
       /* @__PURE__ */ jsx6(P, { children: "Push extracted, resolved, and reviewed data to any downstream system. Delivery is a typed, at-least-once pipeline with idempotency keys on the wire, append-only history, and a dead-letter queue for terminal failures." }),
       /* @__PURE__ */ jsx6(SubHeading, { id: "delivery-pipeline", children: "How Delivery Works" }),
@@ -6483,6 +6536,7 @@ var PLATFORM_NAV_SECTIONS = [
     { id: "field-registry", label: "Field Registry" },
     { id: "tier-system", label: "Tier System" },
     { id: "semantic-clusters", label: "Semantic Clusters" },
+    { id: "field-resolution", label: "Field Resolution" },
     { id: "master-instructions", label: "Master Instructions" }
   ] },
   { id: "schemas-templates", label: "Schemas & Templates", children: [
@@ -6510,10 +6564,15 @@ var PLATFORM_NAV_SECTIONS = [
     { id: "cases", label: "Cases" },
     { id: "document-graph", label: "Document Graph" }
   ] },
+  { id: "data-products", label: "Data Products", children: [
+    { id: "dataset-templates", label: "Dataset Templates" },
+    { id: "assemblies", label: "Assemblies" }
+  ] },
   { id: "validation-quality", label: "Validation & Quality", children: [
     { id: "validation-checks", label: "Validation Checks" },
     { id: "ground-truth", label: "Golden Samples" },
-    { id: "approval-gates", label: "Approval Gates" }
+    { id: "approval-gates", label: "Approval Gates" },
+    { id: "approval-queue", label: "Approval Queue" }
   ] },
   { id: "delivery", label: "Outputs", children: [
     { id: "destinations", label: "Destinations" },
@@ -6576,7 +6635,8 @@ var PLATFORM_SECTION_META = [
   { id: "structuring", title: "Structuring", description: "Vocabulary-driven structuring system with seven field strategies, four processing tiers, capture sub-moves, modifiers, constraints, and synthesize fallback." },
   { id: "extraction-jobs", title: "Extraction Jobs", description: "The 4-phase pipeline: Resolve, Agent, Validation, Re-read. Progressive grid filling with per-cell provenance and confidence." },
   { id: "linking-cases", title: "Linking & Cases", description: "Document linking discovers relationships through shared entities. Related documents grouped into cases with evidence chains." },
-  { id: "validation-quality", title: "Validation & Quality", description: "Automated validation checks, golden sample benchmarking, and configurable approval gates for quality assurance." },
+  { id: "data-products", title: "Data Products", description: "Dataset templates for reusable output specifications and assemblies for combining multi-source documents into structured datasets." },
+  { id: "validation-quality", title: "Validation & Quality", description: "Automated validation checks, golden sample benchmarking, configurable approval gates, and a manual review queue for flagged results." },
   { id: "delivery", title: "Outputs & Delivery", description: "Push structured data to webhooks, REST APIs, SFTP, email, or cloud storage with field mappings, triggers, delivery configuration, and telemetry." },
   { id: "workspace-settings", title: "Workspace Settings", description: "Shared dialects for output formatting, reference primitives for lookup tables, and change review governance for production workspaces." },
   { id: "search-filtering", title: "Search & Filtering", description: "Global omnisearch (Cmd+K) and advanced document filtering by extracted field values with saved presets." },