26 tools with consistent naming (verb_noun pattern) and complete schemas. Descriptions are present but generic (avg ~60 chars, below 194-char baseline). All parameters have type definitions and descriptions. Output schemas not explicitly documented in source. Error handling returns isError flag but lacks recovery guidance. No tool annotations (readOnlyHint/destructiveHint). Composition is sound, tools are single-responsibility and chainable via ID fields. Security: credentials injected via env vars (good), but no permission gates or audit logging visible.

• Expand tool descriptions to 100 - 200 chars. Include WHEN to call (e.g., 'Call this to discover available spaces before creating a page'), WHAT it returns (e.g., 'Returns paginated list with space ID, name, and member count'), and any prerequisites (e.g., 'Requires read:spaces permission').
• Add tool annotations: mark all read-only tools with readOnlyHint=true, all destructive tools with destructiveHint=true, and idempotent operations with idempotentHint=true. This enables LLMs to reason about safety and retry logic.
• Enhance error messages with recovery guidance. Instead of 'Error: Space not found', return 'Space not found. Try docmost_search() or docmost_list_spaces() to find the correct space ID.'
• Document output schemas explicitly in tool descriptions or a separate schema registry. Specify fields, types, and pagination structure (e.g., 'Returns {spaces: [{id, name, description}], total: number, page: number}').
• Implement permission checks and audit logging for all write/delete operations. Log caller identity, operation, parameters, timestamp, and result to enable compliance and incident investigation.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry