Intelligent Scientific Reasoning through Graph-of-Thoughts MCP Server
Critical gaps across all dimensions. Tool 'initialize' violates MCP protocol (should not be a tool). 'asr_got.query' has minimal description (67 chars, below 100-char baseline) and no output schema documented. 'shutdown' lacks description entirely. No parameter descriptions for any tool. No error handling guidance. No tool annotations. Schemas present but incomplete, missing type constraints, enums, and validation rules. Code shows FastAPI HTTP server but tool definitions are inferred from JSON-RPC handler, not explicit MCP registration.
Process a scientific reasoning query using Graph-of-Thoughts methodology
Initialize the MCP server and return server metadata
Shutdown the MCP server
Tool 'initialize' should not exist as a tool. MCP protocol handles server initialization via the initialize request, not a callable tool. This violates the MCP spec and confuses clients.
No parameter descriptions. 'asr_got.query' accepts 'query' but provides no guidance on format, length, or expected content. LLMs cannot infer parameter meaning from names alone.
Tool 'shutdown' has no description. LLMs cannot determine when or why to call it, and cannot distinguish it from other state-modifying tools.
No output schemas documented. 'asr_got.query' returns unstructured reasoning traces and graph state, but the response structure is not formally defined. LLMs cannot plan downstream calls or extract fields reliably.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 26 | <=2025-11-25 | v2 |
No error handling guidance. Code shows generic HTTP exception handling but provides no recovery hints. LLMs receive raw errors with no actionable next steps.