Turn an OpenAPI spec or any MCP server into a Slack, LINE, or Discord chatbot, with in-chat OAuth, human-in-the-loop approval, and surface-native rich replies.
Single tool 'revoke_authorization' has a clear, descriptive name and good description (127 chars), but the input schema is completely empty ({}). The tool lacks parameter descriptions, type definitions, and output schema documentation. No error handling guidance. Tool is write-risk but lacks confirmation/dry-run pattern. Overall definition quality is poor due to schema absence and missing output documentation.
Sign the user out by revoking their authorization for every connected service. Call this only when the user asks to sign out, log out, disconnect, switch accounts, or revoke access. It deletes the stored OAuth tokens, so the next action re-authorizes.
No output schema documented. LLM cannot predict what revoke_authorization returns (success boolean? error details? token list?). Breaks tool-chaining and response parsing.
Destructive operation (revoke_authorization) lacks confirmation/dry-run pattern. Agents can accidentally sign out users without approval. No idempotent or destructive hint annotations visible.
No error handling guidance. Tool description does not explain failure modes (e.g., 'no tokens to revoke', 'service unavailable') or recovery steps for LLM.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 34 | 2026-07-28+ | v2 |
Tool is single-purpose but server appears to be a chatbot/agent framework. Only one tool exposed suggests incomplete tool inventory or tool definitions not visible in provided source.