Serve Custom Tools via OpenAPI and MCP for Function-Calling LLMs
Three tools with complete schemas and basic descriptions. All tools have proper JSON Schema with typed parameters and descriptions. However, descriptions are minimal (10-30 chars), falling below the 50-200 char LLM-optimized baseline. No error handling guidance, no output schema documentation, and no indication of state-modifying behavior. Tools are simple arithmetic/greeting operations with READ_ONLY risk classification, but descriptions lack context about when/why to use them or what they return.
Add two numbers together.
Greet someone by name.
Multiply two numbers.
Descriptions are too brief (10-30 chars). LLM-optimized descriptions should be 50-200 chars, explaining WHAT the tool does, WHEN to use it, and WHAT it returns. Current descriptions lack context for tool selection.
No output schema documented. LLMs need to know what fields to expect in responses to plan downstream calls and extract data correctly. Add return type documentation for each tool.
No error handling guidance. Error responses should tell LLMs what to do next (retryable, user-fixable, or fatal). Current implementation provides no recovery paths.
Tool names lack action verbs. 'add', 'greet', 'multiply' are acceptable but generic. Consider 'calculate_sum', 'send_greeting', 'calculate_product' for clarity in multi-tool contexts.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |