UniFi Announcer has 7 tools with basic descriptions and input schemas, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (auth_check, announce_command, buzzer_command), which is good. However, parameter descriptions are minimal or missing entirely. For example, 'announce_command' has parameters like 'volume' and 'repeat_times' with only type info (integer) but no range constraints (0-100 is mentioned in description but not in schema). Output schemas are not documented anywhere in the provided code. Error handling is absent, no guidance on what happens when a target is invalid or TTS synthesis fails. The server uses FastAPI with HTTP transport (good), but tool definitions appear to be inferred from route handlers rather than explicitly registered with MCP tool schemas.

• Add JSON Schema constraints (minimum, maximum, enum) to all numeric and categorical parameters. E.g., 'volume': {type: 'integer', minimum: 0, maximum: 100}.
• Document output schemas for all tools. Specify what fields are returned, their types, and what IDs/references are included for chaining. E.g., announce_command should return {job_id: string, status: string, target: string}.
• Expand parameter descriptions to include discovery hints and error cases. E.g., 'target: Target chime or group name. Call list_targets() first to see available names. Names are case-sensitive.'
• Add error handling guidance to tool descriptions. E.g., 'If target is not found, returns 404 with available targets. If TTS synthesis times out, returns 408 with retry guidance.'
• Explicitly register tools with MCP ToolDefinition structures including name, description, inputSchema, and outputSchema. Ensure tool definitions are visible in the MCP server's tools list endpoint.