Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. discover-abilities and get-ability-info are read-only discovery tools with minimal parameter requirements. execute-ability is a write tool with confirmation logic but lacks detailed output schema documentation. No tool descriptions explain what data structures are returned or how to chain tools together. Parameter descriptions are present but sparse (e.g., '_' parameter in discover-abilities is marked 'Unused' rather than explained). No evidence of error handling guidance, recovery paths, or validation rules in descriptions.

• Document the output schema for each tool. Example: 'discover-abilities returns {"abilities": [{"name": string, "description": string, "category": string}], "total": number}'. Include field types and what each field means.
• Replace the '_' parameter in discover-abilities with a proper optional parameter or remove it entirely. If it must exist, explain why in the description and provide a valid example value.
• Add error handling guidance to execute-ability description: 'If the ability fails, the response will include an error code and message. Common errors: PERMISSION_DENIED (user lacks capability), ABILITY_NOT_FOUND (name is invalid), EXECUTION_FAILED (ability raised an exception). Retry with a different ability or ask the user for clarification.'
• Clarify tool chaining in get-ability-info description: 'Call discover-abilities first to list all available ability names. Copy the exact name (case-sensitive) and pass it to this tool to retrieve its full schema. Then use execute-ability with the parameters matching that schema.'
• Document the confirmation_reason field: 'A brief, user-facing question (1-2 sentences) asking for approval before executing the ability. Example: "May I update this post now?" This is shown to the user for confirmation before the ability runs.'