Loading ToolBench
Single tool 'write_note' has a basic schema and minimal description. The tool name lacks a clear action verb (should be 'create_note' or 'add_note'). Description is only 20 characters, below the 34-char p10 baseline for production tools. Input schema is present but parameter 'content' lacks validation constraints (no length limits, format hints, or markdown format specification despite claiming markdown support). No output schema documented. No error recovery guidance. Tool performs a write operation but description does not explicitly state this is a state-modifying action. No pagination, no idempotency guarantees, no rate-limit guidance.
Write note to flomo
Tool name 'write_note' is not a clear action verb. Should be 'create_note' or 'add_note' to match verb_noun convention and help LLMs infer intent from the name alone.
Description 'Write note to flomo' is only 20 characters, below production baseline (p10=34 chars). Lacks context on WHEN to use, WHAT it returns, and that it modifies state. LLMs cannot determine tool selection from this minimal text.
Parameter 'content' has no validation constraints. Description claims 'markdown format' but does not specify length limits, character restrictions, or format validation rules. LLMs cannot infer constraints from names alone.
No output schema documented. Response structure (flomoUrl, success message) is not formally defined. LLMs cannot plan downstream tool calls or extract the memo_id/slug for reference without guessing the response format.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 32 | 2026-07-28+ | v2 |
Error handling returns raw error messages ('Failed to write note to flomo: unknown error') without recovery guidance. LLM does not know whether to retry, ask user, or abort. No categorization of retryable vs fatal errors.