An MCP server that provides interactive user input capabilities through command windows and intensive chat sessions, with desktop notifications support.
Server defines 5 tools with basic schemas and descriptions, but exhibits critical gaps across naming, parameter documentation, and output specification. Tool names lack clear action verbs (e.g., 'message_complete_notification' is passive; 'start_intensive_chat' and 'stop_intensive_chat' are vague about what 'intensive' means). Descriptions are present but generic (avg ~60 chars, below 194-char baseline). Parameters lack type constraints and detailed format guidance. No output schemas documented. Error handling is absent, no recovery guidance, retryability classification, or actionable error messages. The tool set conflates UI interaction concerns (request_user_input, message_complete_notification) with session management (start/ask/stop_intensive_chat), violating single-responsibility principle.
Ask a question within an active intensive chat session
Send a desktop notification when a message is complete
Request user input through a command window prompt with optional timeout and predefined options
Start a new intensive chat session for multi-turn conversations with the user
Stop an active intensive chat session
Tool names lack action verbs or are passive/vague. 'message_complete_notification' is passive (should be 'send_notification'). 'start_intensive_chat' and 'stop_intensive_chat' use vague adjective 'intensive', unclear what distinguishes this from a regular chat. LLMs cannot infer intent from these names.
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields. E.g., 'start_intensive_chat' presumably returns a sessionId, but this is not specified in the definition.
Parameter descriptions lack format constraints and validation rules. 'predefinedOptions' is an array of strings with no guidance on max length, character restrictions, or examples. 'sessionId' has no format hint (UUID? hex? length?). LLMs will guess and pass invalid values.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 41 | <=2025-11-25 | v2 |
No error handling or recovery guidance. Tools do not document what happens on timeout, invalid sessionId, or user cancellation. LLMs have no guidance on retryability or next steps.
Tool set violates single-responsibility principle. UI interaction (request_user_input, message_complete_notification) is mixed with session management (start/ask/stop_intensive_chat). These should be separate, composable tools with clear boundaries.