Critical gaps across all dimensions. Tool definitions exist but lack essential structure. No visible input schemas in source code (apps/api/src/interfaces/http/mcp/tools.rs not provided in detail). Descriptions are present but minimal (10-50 chars), below the 194-char baseline for production tools. No parameter documentation visible. No output schema documentation. All 6 tools appear to be READ_ONLY, which is positive for safety, but the definitions themselves are severely underdeveloped for agent use.

• Add explicit JSON Schema input definitions for all tools. Include type, description, enum constraints, and min/max for every parameter. Example: { 'workspace_id': { 'type': 'string', 'description': 'The workspace ID or name to query. If you only have a name, call list_workspaces() first.' } }
• Expand tool descriptions to 100-200 chars. Include WHAT the tool does, WHEN to use it, and what structure it returns. Example: 'List all available workspaces in the IronRAG instance. Call this first to discover workspace IDs before searching documents. Returns array of {workspace_id, name, created_at}.'
• Document output schemas for all tools. Specify field names, types, and whether they are required. Example for list_workspaces: 'Returns {workspaces: [{id: string, name: string, doc_count: number}], total: number}'
• Add pagination to list_* tools: accept limit (1-100, default 20) and offset/cursor parameters. Return total_count and next_cursor in response to enable agent iteration over large result sets.
• Add parameter descriptions for search_documents and read_document. Clarify: what is the search query syntax? What fields does read_document return? Does view_document_image require a page number?