get_tool

Inspect a named tool after search_tools. Set detail=‘full’ for its exact schema and complete authored description; carry workspace_id for cross-workspace calls. For run_model, full detail also returns the canonical output-lineage authoring contract.

read-only · idempotent

Parameters

Name	Type	Required	Default	Description
name	Annotated[str, Field(description="Name of the MCP tool to look up (e.g. 'run_model', 'create_file', 'query_data').")]	Yes	—
detail	Annotated[ToolDetail, Field(description='Detail level to return: schema, usage, or full.')]	No	'schema'

Returns

Returns a structured object. Top-level fields:

• name (string)
• title (string)
• domain (string)
• tier (string)
• description (string)
• full_description (string, when detail='full')
• input_schema (object)
• output_schema (null)
• annotations (object)
• usage_guidance (string, when detail='usage' or detail='full')
• required_role (string or null)
• required_scopes (array<string>)
• callable (boolean or null)
• related_tools (array<string>)
• supplemental_contracts (object, when detail='full' and the tool has a structured contract; run_model currently exposes output_lineage)

See the response example below for the full payload shape.

For get_tool(name="run_model", detail="full"), supplemental_contracts.output_lineage contains the current and accepted schema versions, required artifact fields, /outputs/output_lineage.json, BT_RUN_ID, coordinate-coverage options, the manifest-backed helper release and checksum, and a minimal Python/artifact example.

Example

Tool Call

{
  "name": "get_tool",
  "arguments": {
    "name": "create_dashboard_from_run"
  }
}

Response

{
  "name": "create_dashboard_from_run",
  "title": "Create Dashboard From Run",
  "domain": "dashboard",
  "tier": "cold",
  "description": "Create a dashboard from one successful model-run output.",
  "input_schema": {
    "type": "object",
    "required": [
      "model_name",
      "dashboard_name",
      "chart_type",
      "model_run_id",
      "output_key"
    ],
    "properties": {
      "model_name": {
        "type": "string"
      },
      "dashboard_name": {
        "type": "string"
      },
      "chart_type": {
        "type": "string"
      },
      "model_run_id": {
        "type": "string"
      },
      "output_key": {
        "type": "string"
      },
      "description": {
        "type": "string"
      },
      "subtitle": {
        "type": "string"
      },
      "currency": {
        "type": "string"
      }
    }
  },
  "output_schema": null,
  "annotations": {
    "title": "Create Dashboard From Run",
    "readOnlyHint": false,
    "destructiveHint": true,
    "idempotentHint": false,
    "openWorldHint": false
  },
  "related_tools": [
    "publish_repo_dashboard",
    "manage_dashboard_share",
    "list_dashboards"
  ]
}

Notes

• Module: Tool Discovery
• Requires an authenticated Bridge Town session (OAuth) or an API token.