This server has fundamental definition quality issues that prevent production use. The single tool 'flomo' has a Chinese-language description with no English alternative, missing parameter descriptions, no output schema documentation, and incomplete error handling guidance. The tool name lacks an action verb and does not follow standard naming conventions. The implementation has hardcoded empty credentials (flomo_api = '') and poor input validation. No evidence of following agentic tool patterns.
记录到flomo
Tool name 'flomo' lacks action verb, should be 'create_flomo_memo' or 'save_to_flomo'. LLMs rely on verb-noun naming to infer intent before reading description.
Tool description is in Chinese ('记录到flomo') with no English translation. LLMs in English-speaking contexts cannot understand intent. Production tools must have clear English descriptions.
The sole input parameter 'content' has a description that is borderline adequate ('The content to record to Flomo') but lacks format constraints, length limits, or examples of valid input. No guidance on special handling (markdown, links, tags).
No documented output schema. The code returns text-only responses, but the internal flomoResponse interface shows the API returns structured data (creator_id, tags, created_at, linked_count). LLMs have no visibility into what fields to expect or how to chain downstream calls.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 24 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 17 | - | v1 |
Error handling is minimal. The tool returns generic text ('记录失败' / 'Record failed', 'Content cannot be empty') with no recovery guidance. When the API fails with 'code !== 0', the error message from the API is returned but not parsed into actionable next steps for the LLM.
Credentials are not properly injected. The code declares 'let flomo_api = ""' and then uses it unchanged. While dotenv is imported and attempted to load from .env.local, the API endpoint is never actually populated. This will cause all requests to fail silently.
No permission gates or scope declarations. The tool can record arbitrary content to the user's Flomo account, a destructive operation that modifies user data. No checks for authorization or scope validation.
No audit logging or trace of who called the tool, when, and with what content. Agent actions that modify user data must be logged for compliance and debugging.