Flanner exposes 4 tools with clear verb-noun naming (get_, list_, create_, initialize_). All tools have descriptions (avg 95 chars, within baseline 34-392). Input schemas are present with type declarations. However, parameter descriptions are sparse or missing entirely, most params lack actionable guidance on format, constraints, or dependencies. Output schemas are not documented. Error handling and recovery guidance are absent. The server follows basic MCP patterns but lacks the depth expected of production-grade tools.

• Document output schemas for all tools. For list_projects, specify: returns array of {project_id: string, name: string, description: string, project_root: string, plan_directory: string}. For create_project_tool, specify: returns {project_id: string, name: string, created_at: ISO8601}.
• Expand parameter descriptions with format and validation rules. E.g., project_id: 'UUID string (e.g., 550e8400-e29b-41d4-a716-446655440000). If omitted, returns default configuration.' project_root: 'Absolute filesystem path to project root. Must be a valid directory. If omitted, auto-detected from current working directory.'
• Add error handling guidance to tool descriptions. E.g., create_project_tool: 'Returns error if project name already exists (try list_projects to check), if project_root is invalid, or if git is not initialized in that directory.'
• Add pagination to list_projects: include limit (default 20, max 100) and offset parameters. Return {projects: [...], total: number, offset: number} so agents can iterate over large project lists.
• Clarify idempotency in tool descriptions. E.g., create_project_tool: 'Idempotent: calling twice with the same name and root returns the same project_id without creating a duplicate.' initialize_project_tool: 'Idempotent: safe to call multiple times on the same repo; skips steps already completed.'

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

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