Repository containing multiple Kanban board applications and portfolio projects with AI integration
This is a collection of frontend/backend project files, not an MCP server. No MCP protocol implementation detected. Tool definitions appear to be inferred from FastAPI endpoints and Next.js route handlers, not explicit MCP tool registrations. Descriptions are present but minimal (10-50 chars). Input schemas are partially visible but lack proper JSON Schema structure with type definitions. No error handling guidance, no output schemas documented, no tool annotations. Transport mechanism unknown, likely HTTP but not MCP-compliant.
Tests AI integration by calling OpenRouter API with a simple math question (2+2)
Deletes a card from the board by card ID
AI digital twin of Qasim Hassan that answers questions based on his career profile using OpenRouter API
Retrieves the current Kanban board state
Returns a hello world message
Authenticates user with username and password, sets HTTP cookie with access token
Logs out user by deleting authentication cookie
No MCP protocol implementation detected. Source code shows FastAPI/Next.js projects, not an MCP server. Tool definitions are inferred from HTTP endpoints, not explicit MCP tool registrations.
Input schemas missing or incomplete for 7 of 12 tools (hello, logout, me, ai_test, get_board). No JSON Schema type definitions visible. Cannot verify parameter constraints.
Descriptions are too short (10-50 chars) for 8 tools. Baseline for A+ tools is 50-200 chars. Descriptions lack WHEN to use, WHAT it returns, and prerequisites. Examples: 'Returns a hello world message' (hello), 'Logs out user by deleting authentication cookie' (logout).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | 2026-07-28+ | v2 |
Returns the authenticated user's username
Updates a card's properties (title, details, column, or position)
Updates a column's title by column ID
Creates a new card in a specified column
Replaces the entire Kanban board with new state
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields. No pagination support visible for list-like operations (get_board).
No error handling guidance. No recovery hints, no categorization of retryable vs fatal errors, no actionable error messages. Destructive operations (delete_card) lack confirmation or dry-run support.