@salesforce/afv-skills 1.64.0 → 1.65.0

package/skills/data360-intelligent-document-processing-create/SKILL.md

@@ -0,0 +1,148 @@
+---
+name: data360-intelligent-document-processing-create
+description: "Use this skill when users want to execute Data Cloud IntelligentDocumentProcessing operations via the Data 360 MCP facade. Trigger when users mention data360/intelligent, document, processing, shared, cdp, mcp, connect, api, shared-cdp-mcp-connect-api."
+metadata:
+  version: "1.0"
+---
+
+# Managing Data360/intelligent Document Processing
+
+## When to Use This Skill
+
+Execute Data Cloud IntelligentDocumentProcessing operations via the Data 360 MCP facade.
+
+## Workflow
+
+Each step lists an `operation` id. Call describe(<operation>) BEFORE dispatching to get that tool's parameter schema and, where available, a worked example. `paramsJson` decodes to the operation's requestBody schema VERBATIM, including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap or omit it. The only envelope the step handles for you is `{toolName, paramsJson}` (its `invocation` already wraps it); do not POST the operation path directly. If the tool's error suggests `payload_examples`, ignore it — that tool isn't available here; call describe(<operation>) instead.
+
+## Critical Constraints
+
+**Destructive operations — require explicit user confirmation:**
+
+- **`document-ai-configuration-delete`** — Delete a Document AI configuration (internally named IDP configuration) by its ID or API name. This permanently removes the configuration and its associated schema and rules.
+  - Rollback: No automatic rollback.
+
+## Operations Reference
+
+Operations grouped by purpose. Use these as the building blocks for the workflows above.
+
+### Summary
+
+| Operation | Purpose | Status | Call | Depends on |
+|-----------|---------|--------|------|------------|
+| `document-ai-configuration-create` | write | implemented | `execute_document-ai-configuration-create` | — |
+| `document-ai-configuration-delete` | delete | implemented | `execute_document-ai-configuration-delete` | — |
+| `document-ai-configuration-get` | read | implemented | `execute_document-ai-configuration-get` | — |
+| `document-ai-configuration-list` | read | implemented | `execute_document-ai-configuration-list` | — |
+| `document-ai-configuration-process` | write | implemented | `execute_document-ai-configuration-process` | — |
+| `document-ai-configuration-update` | write | implemented | `execute_document-ai-configuration-update` | — |
+| `document-ai-data-extract` | read | implemented | `execute_document-ai-data-extract` | — |
+| `document-ai-global-config-get` | read | implemented | `execute_document-ai-global-config-get` | — |
+| `document-ai-schema-detect` | read | implemented | `execute_document-ai-schema-detect` | — |
+| `document-ai-schema-generate` | write | implemented | `execute_document-ai-schema-generate` | — |
+
+### Read operations
+
+#### `document-ai-configuration-get`
+
+Get the details of a specific Document AI configuration (internally named IDP configuration) by its ID or API name. Returns configuration details including schema, data model objects, and processing rules.
+
+**Call:** `execute_document-ai-configuration-get`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_get (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-get) for the param schema and, where available, a worked example.
+
+#### `document-ai-configuration-list`
+
+List all Document AI configurations (internally named IDP configurations) with optional filtering by activation status, search text, and pagination support. Returns configuration details including labels, API names, and status.
+
+**Call:** `execute_document-ai-configuration-list`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_list (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-list) for the param schema and, where available, a worked example.
+
+#### `document-ai-data-extract`
+
+Extract structured data from documents by providing a json schema or Document AI configuration. Supports targeted page-range extraction and confidence scores for the extracted data. Returns extracted field values and metadata.
+
+**Call:** `execute_document-ai-data-extract`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_data_extract (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-data-extract) for the param schema and, where available, a worked example.
+
+#### `document-ai-global-config-get`
+
+Get the global configuration settings for Document AI (internally named IDP). Returns system-wide Document AI settings including enabled features, quotas, and default parameters.
+
+**Call:** `execute_document-ai-global-config-get`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_global_config_get (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-global-config-get) for the param schema and, where available, a worked example.
+
+#### `document-ai-schema-detect`
+
+Classify a document against existing Document AI schema configurations (internally named IDP schema configurations) and return the best matching configuration along with a confidence score. This operation identifies which existing schema configuration applies to the provided document and does not generate a new schema. To create a new schema, use d360_document_ai_schema_generate. Required inputs: schemaConfigurations (api names of existing configurations), mlModel, files, and mode. Threshold is optional, defaults if omitted, and must be between 0.0 and 1.0.
+
+**Call:** `execute_document-ai-schema-detect`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_schema_detect (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-schema-detect) for the param schema and, where available, a worked example.
+
+### Write operations
+
+#### `document-ai-configuration-create`
+
+Create a new Document AI configuration (internally named IDP configuration) with schema, data model objects, and processing rules. Specify the source and filter data model objects, ML model, and activation status. Returns the created configuration details. Before calling this tool, the agent MUST: (a) call d360_document_ai_global_config_get to fetch the list of supported models, present that list to the user, and ask which mlModel to use — do not guess or hard-code an mlModel value; and (b) enumerate the supported unstructured DMOs (UDMOs) by calling d360_dmo_list with dataObjectCategory=DirectoryTable, then present that list to the user and ask which UDMO to bind the configuration to (or whether to leave it unbound). Then explicitly ask the user: (1) whether to bind the configuration to a sourceDmoDeveloperName from the enumerated UDMO list (for batch/scheduled processing of files in an existing unstructured DMO) or leave it unbound (for standalone/on-demand extraction)—this choice cannot be changed later; (2) which file/content types the configuration should accept, and provide them as MIME types in fileConfig.fileTypes (for example, application/pdf, image/png, image/jpeg); and (3) whether they want to specify custom output DLO and field names. If yes, collect the required names and pass them in extractedDloConfig. If no, omit extractedDloConfig so the output DLO and field names are automatically derived from the schema.
+
+**Call:** `execute_document-ai-configuration-create`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_create (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-create) for the param schema and, where available, a worked example.
+
+#### `document-ai-configuration-process`
+
+Process (full refresh) a Document AI configuration (internally named IDP configuration) by its ID or API name. This triggers a full reprocessing of all documents associated with the configuration. Returns empty on success.
+
+**Call:** `execute_document-ai-configuration-process`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_process (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-process) for the param schema and, where available, a worked example.
+
+#### `document-ai-configuration-update`
+
+Update a Document AI configuration (internally named IDP configuration) by its ID or API name. Allows modification of configuration details, schema, and processing rules. Returns the updated configuration.
+
+**Call:** `execute_document-ai-configuration-update`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_update (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-update) for the param schema and, where available, a worked example.
+
+#### `document-ai-schema-generate`
+
+Generate a Document AI schema (internally named IDP schema) by analyzing sample documents. Provide document content, optional user description, and context. Returns a generated schema with field definitions.
+
+**Call:** `execute_document-ai-schema-generate`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_schema_generate (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-schema-generate) for the param schema and, where available, a worked example.
+
+### Destructive operations
+
+#### `document-ai-configuration-delete`
+
+Delete a Document AI configuration (internally named IDP configuration) by its ID or API name. This permanently removes the configuration and its associated schema and rules.
+
+**Call:** `execute_document-ai-configuration-delete`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_document_ai_configuration_delete (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_document-ai-configuration-delete) for the param schema and, where available, a worked example.

package/skills/data360-knowledge-graph-construction-create/SKILL.md

@@ -0,0 +1,93 @@
+---
+name: data360-knowledge-graph-construction-create
+description: "Use this skill when users want to execute Data Cloud KnowledgeGraphConstruction operations via the Data 360 MCP facade. Trigger when users mention data360/knowledge, graph, construction, shared, cdp, mcp, connect, api, shared-cdp-mcp-connect-api."
+metadata:
+  version: "1.0"
+---
+
+# Managing Data360/knowledge Graph Construction
+
+## When to Use This Skill
+
+Execute Data Cloud KnowledgeGraphConstruction operations via the Data 360 MCP facade.
+
+## Workflow
+
+Each step lists an `operation` id. Call describe(<operation>) BEFORE dispatching to get that tool's parameter schema and, where available, a worked example. `paramsJson` decodes to the operation's requestBody schema VERBATIM, including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap or omit it. The only envelope the step handles for you is `{toolName, paramsJson}` (its `invocation` already wraps it); do not POST the operation path directly. If the tool's error suggests `payload_examples`, ignore it — that tool isn't available here; call describe(<operation>) instead.
+
+## Critical Constraints
+
+**Destructive operations — require explicit user confirmation:**
+
+- **`knowledge-graph-construction-delete`** — Delete a Knowledge Graph construction config by name or ID. This permanently 
+  - Rollback: No automatic rollback.
+
+## Operations Reference
+
+Operations grouped by purpose. Use these as the building blocks for the workflows above.
+
+### Summary
+
+| Operation | Purpose | Status | Call | Depends on |
+|-----------|---------|--------|------|------------|
+| `knowledge-graph-construction-create` | write | implemented | `execute_knowledge-graph-construction-create` | — |
+| `knowledge-graph-construction-delete` | delete | implemented | `execute_knowledge-graph-construction-delete` | — |
+| `knowledge-graph-construction-get` | read | implemented | `execute_knowledge-graph-construction-get` | — |
+| `knowledge-graph-construction-list` | read | implemented | `execute_knowledge-graph-construction-list` | — |
+| `knowledge-graph-construction-update` | write | implemented | `execute_knowledge-graph-construction-update` | — |
+
+### Read operations
+
+#### `knowledge-graph-construction-get`
+
+Get a Knowledge Graph construction config by name or ID, including 
+
+**Call:** `execute_knowledge-graph-construction-get`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_construction_get (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-construction-get) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-construction-list`
+
+List construction configs for a Knowledge Graph. 
+
+**Call:** `execute_knowledge-graph-construction-list`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_construction_list (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-construction-list) for the param schema and, where available, a worked example.
+
+### Write operations
+
+#### `knowledge-graph-construction-create`
+
+Create the construction (build) config for a Knowledge Graph: semantic 
+
+**Call:** `execute_knowledge-graph-construction-create`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_construction_create (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-construction-create) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-construction-update`
+
+Update (full replace) an existing Knowledge Graph construction config by name or ID. 
+
+**Call:** `execute_knowledge-graph-construction-update`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_construction_update (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-construction-update) for the param schema and, where available, a worked example.
+
+### Destructive operations
+
+#### `knowledge-graph-construction-delete`
+
+Delete a Knowledge Graph construction config by name or ID. This permanently 
+
+**Call:** `execute_knowledge-graph-construction-delete`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_construction_delete (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-construction-delete) for the param schema and, where available, a worked example.

package/skills/data360-knowledge-graph-create/SKILL.md

@@ -0,0 +1,216 @@
+---
+name: data360-knowledge-graph-create
+description: "Use this skill when users want to execute Data Cloud KnowledgeGraph operations via the Data 360 MCP facade. Trigger when users mention data360/knowledge, graph, shared, cdp, mcp, connect, api, shared-cdp-mcp-connect-api."
+metadata:
+  version: "1.0"
+---
+
+# Managing Data360/knowledge Graph
+
+## When to Use This Skill
+
+Execute Data Cloud KnowledgeGraph operations via the Data 360 MCP facade.
+
+## Workflow
+
+Each step lists an `operation` id. Call describe(<operation>) BEFORE dispatching to get that tool's parameter schema and, where available, a worked example. `paramsJson` decodes to the operation's requestBody schema VERBATIM, including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap or omit it. The only envelope the step handles for you is `{toolName, paramsJson}` (its `invocation` already wraps it); do not POST the operation path directly. If the tool's error suggests `payload_examples`, ignore it — that tool isn't available here; call describe(<operation>) instead.
+
+## Critical Constraints
+
+**Destructive operations — require explicit user confirmation:**
+
+- **`knowledge-graph-delete`** — Delete a Knowledge Graph by name or ID. This permanently removes the graph 
+  - Rollback: No automatic rollback.
+- **`knowledge-graph-extraction-delete`** — Delete a Knowledge Graph extraction definition by name. This permanently 
+  - Rollback: No automatic rollback.
+
+## Operations Reference
+
+Operations grouped by purpose. Use these as the building blocks for the workflows above.
+
+### Summary
+
+| Operation | Purpose | Status | Call | Depends on |
+|-----------|---------|--------|------|------------|
+| `knowledge-graph-build` | write | implemented | `execute_knowledge-graph-build` | — |
+| `knowledge-graph-build-status` | read | implemented | `execute_knowledge-graph-build-status` | — |
+| `knowledge-graph-create` | write | implemented | `execute_knowledge-graph-create` | — |
+| `knowledge-graph-delete` | delete | implemented | `execute_knowledge-graph-delete` | — |
+| `knowledge-graph-extraction-create` | write | implemented | `execute_knowledge-graph-extraction-create` | — |
+| `knowledge-graph-extraction-delete` | delete | implemented | `execute_knowledge-graph-extraction-delete` | — |
+| `knowledge-graph-extraction-get` | read | implemented | `execute_knowledge-graph-extraction-get` | — |
+| `knowledge-graph-extraction-list` | read | implemented | `execute_knowledge-graph-extraction-list` | — |
+| `knowledge-graph-extraction-update` | write | implemented | `execute_knowledge-graph-extraction-update` | — |
+| `knowledge-graph-get` | read | implemented | `execute_knowledge-graph-get` | — |
+| `knowledge-graph-list` | read | implemented | `execute_knowledge-graph-list` | — |
+| `knowledge-graph-presign-urls` | read | implemented | `execute_knowledge-graph-presign-urls` | — |
+| `knowledge-graph-replace` | write | implemented | `execute_knowledge-graph-replace` | — |
+| `knowledge-graph-suggest-smart-config` | read | implemented | `execute_knowledge-graph-suggest-smart-config` | — |
+| `knowledge-graph-suggest-smart-config-status` | read | implemented | `execute_knowledge-graph-suggest-smart-config-status` | — |
+| `knowledge-graph-update` | write | implemented | `execute_knowledge-graph-update` | — |
+
+### Read operations
+
+#### `knowledge-graph-build-status`
+
+KG Playground, step 6 poll: get the status of the current playground (REALTIME) 
+
+**Call:** `execute_knowledge-graph-build-status`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_build_status (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-build-status) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-extraction-get`
+
+Get a Knowledge Graph extraction definition by name. 
+
+**Call:** `execute_knowledge-graph-extraction-get`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_extraction_get (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-extraction-get) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-extraction-list`
+
+List Knowledge Graph extraction definitions, with paging (limit, offset). 
+
+**Call:** `execute_knowledge-graph-extraction-list`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_extraction_list (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-extraction-list) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-get`
+
+Get a Knowledge Graph by name or ID, including the resolved (suffixed) 
+
+**Call:** `execute_knowledge-graph-get`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_get (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-get) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-list`
+
+List Knowledge Graphs in the org, with paging (limit, offset). 
+
+**Call:** `execute_knowledge-graph-list`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_list (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-list) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-presign-urls`
+
+KG Playground, step 1 (REALTIME/playground graphs only): request presigned upload 
+
+**Call:** `execute_knowledge-graph-presign-urls`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_presign_urls (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-presign-urls) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-suggest-smart-config`
+
+KG Playground, step 4 (submit): kick off an async AI recommendation that returns an 
+
+**Call:** `execute_knowledge-graph-suggest-smart-config`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_suggest_smart_config (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-suggest-smart-config) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-suggest-smart-config-status`
+
+KG Playground, step 4 poll: get the status/result of the current 
+
+**Call:** `execute_knowledge-graph-suggest-smart-config-status`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_suggest_smart_config_status (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-suggest-smart-config-status) for the param schema and, where available, a worked example.
+
+### Write operations
+
+#### `knowledge-graph-build`
+
+KG Playground, step 6 (final playground step): trigger an async build of a 
+
+**Call:** `execute_knowledge-graph-build`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_build (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-build) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-create`
+
+Create a Data Cloud Knowledge Graph in a data space. FIRST ASK THE USER: 
+
+**Call:** `execute_knowledge-graph-create`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_create (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-create) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-extraction-create`
+
+Create an extraction (LLM mining job) that pulls entities and relationships 
+
+**Call:** `execute_knowledge-graph-extraction-create`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_extraction_create (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-extraction-create) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-extraction-update`
+
+Update (full replace) an existing Knowledge Graph extraction definition by name. 
+
+**Call:** `execute_knowledge-graph-extraction-update`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_extraction_update (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-extraction-update) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-replace`
+
+Replace (full update) an existing Knowledge Graph by name or ID. PUT 
+
+**Call:** `execute_knowledge-graph-replace`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_replace (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-replace) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-update`
+
+Update an existing Knowledge Graph by name or ID (partial update of its 
+
+**Call:** `execute_knowledge-graph-update`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_update (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-update) for the param schema and, where available, a worked example.
+
+### Destructive operations
+
+#### `knowledge-graph-delete`
+
+Delete a Knowledge Graph by name or ID. This permanently removes the graph 
+
+**Call:** `execute_knowledge-graph-delete`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_delete (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-delete) for the param schema and, where available, a worked example.
+
+#### `knowledge-graph-extraction-delete`
+
+Delete a Knowledge Graph extraction definition by name. This permanently 
+
+**Call:** `execute_knowledge-graph-extraction-delete`   — status: `implemented`
+
+**Inputs:**
+
+- `paramsJson` *(`string`)* — JSON-encoded params for d360_knowledge_graph_extraction_delete (a STRING, double-encoded). Serialize the operation's requestBody schema VERBATIM including its top-level key (e.g. `input`, `dataTransform`) — do not unwrap it. Call describe(execute_knowledge-graph-extraction-delete) for the param schema and, where available, a worked example.