This server has 2 tools with basic schema definitions but significant gaps in descriptions, parameter documentation, and output specifications. Tool names are verb-first but descriptions are minimal (generic/placeholder). Schemas use Zod but lack detail. No output documentation. Error handling is present but generic. This is a typical early-stage MCP server with foundational elements but incomplete for production use.
Tool for getting user information
Example tool structure - will be replaced with actual implementations
Tool descriptions are below 20 characters minimum and lack actionable context. 'Tool for getting user information' (32 chars) is generic; 'Example tool structure - will be replaced with actual implementations' is a placeholder, not a real description.
No output schemas documented. LLMs cannot plan downstream calls or know what fields to extract from responses. Current code returns generic {content: [{type: 'text', text: ...}]} but does not document structure for agent reasoning.
send-message tool is functionally incomplete (placeholder implementation). Description says 'will be replaced with actual implementations', indicating this is stub code, not production-ready. The WRITE risk classification suggests this tool needs careful error handling and confirmation flows, currently absent.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 35 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 41 | - | v1 |
Parameter descriptions lack actionable guidance. 'Receiver's ID', 'Message type', 'Message content' do not explain format, constraints, or valid values. For msg_type, the description mentions 'text, post, image, etc.' as examples but these should be defined as enums or a regex pattern.
No pagination or limit controls for get-user-info. If the underlying API returns multiple fields or nested structures, there is no way to constrain results. Output could be verbose and waste context.
Error responses are generic (catch-all error message) and do not guide LLM recovery. 'Error fetching user info: <message>' or 'Error: <message>' do not tell the agent whether to retry, ask the user, or try an alternative tool.
send-message tool name is ambiguous. Does 'send-message' send to a user, channel, or group? The parameter 'receive_id' is opaque (unclear whether it's a user_id, channel_id, or group_id). Naming should be more specific (e.g., 'send_user_message') or parameter should be renamed to clarify type.