toyMCP has 4 tools with minimal definition quality. Tool descriptions are present but extremely brief (10-30 chars), well below the 50-200 char baseline for LLM-optimized descriptions. Input schemas are visible but lack parameter descriptions entirely, every parameter is undescribed, violating the critical rule that 'every parameter needs a description.' No output schemas are documented. Error handling is absent. The tools are simple (todo CRUD + discovery), but the definitions lack the depth needed for reliable LLM invocation.

• Add 50-150 character descriptions to each tool explaining WHAT it does, WHEN to use it, and what it returns. Example: 'todo.add: Create a new todo item and return its ID. Use this to add tasks to the list. Returns {id, text, created_at}.'
• Add descriptions to all parameters. For 'text', write: 'The todo item text (1-500 characters, plain text only).' For 'id', write: 'The numeric ID of the todo to remove (obtained from todo.list).'
• Document output schemas for all tools. Example for todo.list: 'Returns {todos: [{id: number, text: string, created_at: ISO8601}], total: number}'.
• Add error handling with recovery hints. Example: 'If todo.remove returns 404, call todo.list() to see available IDs.'
• Rename mcp.discover to list_available_tools or get_tool_schemas for clarity.