assetcore_docs_search

AssetCore is a deterministic structured memory substrate. It stores all world state explicitly in containers and modifies that state through atomic, replayable transactions. Retrieve role-aware reasoning documentation sections (Reasoning, Decision, Pattern, Ontology) with deterministic heading-based scoring.

Details

AssetCore is a deterministic structured memory substrate. It stores all world state explicitly in containers and modifies that state through atomic, replayable transactions. Retrieve role-aware reasoning documentation sections (Reasoning, Decision, Pattern, Ontology) with deterministic heading-based scoring.

Searches the four canonical reasoning documents with role-aware section scoring: Reasoning (conceptual), Decision (tool selection), Pattern (recipes), and Ontology (entity/container definitions). Sections are bounded by Markdown ##/### headings for precise retrieval. The Reasoning Guide supplies conceptual grounding for reasoning queries that need deterministic world-model context. Parameters: query (required string) and max_sections (optional integer, default 3, max 10). Results include doc_id, title, doc_role, heading, raw Markdown content, covered-doc summary, and suggested follow-ups. Output ordering is deterministic with stable tie-breaks.

Protocols

• MCP
• OpenAI
• Gemini

Input Schema

{
  "properties": {
    "max_sections": {
      "default": 3,
      "description": "Maximum number of sections to return (default: 3, max: 10)",
      "maximum": 10,
      "minimum": 1,
      "type": "integer"
    },
    "query": {
      "description": "Natural-language question or topic to retrieve documentation for.",
      "type": "string"
    }
  },
  "required": [
    "query"
  ],
  "type": "object"
}

Examples

No examples are currently available for this tool.