Minimal Yes/No HTTP server for Render free plan, to be called from ChatGPT Apps SDK. Provides cryptographically random yes/no responses via MCP protocol.
Single tool 'yesno' has a basic description (47 chars) and a properly typed input schema with one required parameter. However, the description is minimal and lacks context about when/why to use the tool. No output schema is documented. The tool name lacks a verb prefix (should be 'get_random_answer' or 'flip_coin'). Parameter description is adequate but generic. No error handling guidance or recovery patterns documented. Tool is read-only but this is not explicitly stated in the description.
Return a cryptographically random yes or no.
Tool name lacks action verb. 'yesno' does not follow verb_noun convention (e.g., 'get_random_answer', 'flip_coin'). LLMs rely on verb prefixes to infer intent.
No output schema documented. Tool description does not specify what fields are returned (e.g., {answer: 'yes'|'no'}). LLMs cannot plan downstream operations without knowing response structure.
Tool description is minimal (47 chars, below baseline 194 chars). Lacks context: when should the LLM call this vs. other decision tools? What are use cases? Is it truly random or seeded?
Parameter 'prompt' is marked required but description states it is 'ignored for randomness'. This is confusing, if ignored, why require it? Consider making it optional with a sensible default.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 55 | 2026-07-28+ | v2 |
No error handling or recovery guidance. Tool description does not explain failure modes, retryability, or what to do if the service is unavailable.