A minimal, educational MCP server that provides a single stock snapshot tool using the free Alpha Vantage API.
Single tool with clear naming (get_stock_snapshot) and complete input schema. Description is adequate (145 chars, within 10-1024 range) but lacks actionable context for LLM selection. Schema properly declares required fields, min/max constraints, and additionalProperties=false. However, output schema is completely undocumented, LLMs cannot plan downstream operations or extract fields from the JSON response. No error handling guidance, no recovery paths, and no documentation of what fields the returned snapshot contains.
Return a stock snapshot from Alpha Vantage, including meta info, latest quote, basic fundamentals (if available), and recent daily OHLCV history. Educational / demo use only.
Output schema completely undocumented. Tool returns JSON snapshot but LLMs have no visibility into field structure, types, or what data is available. Prevents downstream tool composition and forces LLMs to guess at response shape.
Tool description lacks WHEN/WHY context. Current description states WHAT (returns snapshot) but not WHEN to call it vs alternatives or what prerequisites exist (API key, rate limits). LLMs cannot determine if this is the right tool for a user's intent.
No error handling guidance. Tool can fail (invalid symbol, API rate limit, missing API key) but provides no recovery hints. LLMs receive raw errors with no actionable next steps.
Parameter descriptions lack format/constraint details. 'history_days' has min/max in schema but description does not explain what happens if omitted (defaults to 60) or why the limit is 100. 'symbol' description uses examples (AAPL, MSFT, TSLA) instead of format constraints.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | <=2025-11-25 | v2 |
No pagination or result-limiting guidance. Tool returns 'recent daily OHLCV history' but does not document max records returned, whether results are paginated, or how to handle large datasets. Could blow context window.