Dark App Factory MCP exposes 7 tools with basic schemas and descriptions, but falls short of production quality. Tool names lack clear action verbs (factory_* prefix is generic; 'factory_fleet' is vague). Descriptions are present but brief (avg ~60 chars, below 194-char baseline). Input schemas are visible and typed, but parameter descriptions are minimal or missing context. Output schemas are undocumented. Error handling and recovery guidance are absent. No evidence of idempotency markers, permission gates, or audit logging. The server is HTTP-based (good), but tool definitions read as inferred from a React UI component rather than explicit MCP registration, suggesting incomplete implementation visibility.

• Rename tools with clear action verbs: 'get_fleet_status', 'start_generation_run', 'poll_run_status', 'cancel_run', 'launch_app', 'analyze_output', 'list_outputs'. Verify names in actual fastmcp server code.
• Add 10 - 50 character descriptions to every parameter. E.g., 'vibe: Plain-language description of the app to build (10 - 8000 chars). Can include a ## Tech Stack section.' Document enum values inline.
• Document output schemas for all tools. E.g., factory_run returns {run_id: string, status: string, output_dir: string}. factory_status returns {runs: [{run_id, status, progress_pct, log_tail}], total: int}.
• Add error handling guidance. E.g., 'If factory_run fails with "invalid vibe", ensure the description is 10+ characters and avoids special characters. If port conflict, try factory_launch with port=null for auto-selection.'
• Verify tool registration in mcp-server/src/dark_app_factory_mcp/server.py. Ensure all 7 tools are explicitly decorated with @mcp.tool() and include full input/output schemas. Provide the actual server.py code for validation.