A Model Context Protocol server that implements a notes system with web content fetching capabilities
Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and error handling. create_note has minimal descriptions (title/content lack context). fetch tool description is adequate but lacks guidance on when to use it vs alternatives, timeout behavior, or error recovery. Neither tool documents output structure for LLM planning. No error handling guidance, agents cannot distinguish retryable vs fatal failures. Parameter descriptions are generic; no constraints on URL format beyond 'format: uri'. Missing idempotency guidance for create_note (could create duplicates on retry). Overall structure is present but falls short of production quality.
Create a new note
Fetch and process content from a URL using Puppeteer
No output schema documentation. LLMs cannot plan downstream tool calls or extract returned data. create_note returns 'Created note {id}: {title}' but schema is undocumented; fetch returns markdown but structure is implicit.
Parameter descriptions lack actionable constraints. 'URL to fetch content from' does not specify timeout, max size, supported protocols, or what happens on 404/403. LLMs cannot validate inputs before calling.
No error handling guidance. Errors like 'Note {id} not found' or fetch timeouts do not tell the LLM what to do next (retry? ask user? try alternative?). Missing recovery paths.
create_note is non-idempotent and lacks confirmation. Retrying with same title/content creates duplicate notes. No dry-run or confirmation step for this write operation.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | <=2025-11-25 | v2 |
fetch tool description mentions Puppeteer but does not explain when to use it vs simpler HTTP fetch, or what content types it handles. Missing dependency hints and use-case guidance.