MCP Server for extracting and querying system logs with Sentry and Datadog integration
Five tools with reasonable naming (verb_noun pattern) and descriptions (100-200 chars), but critical gaps in schema completeness and parameter documentation. All tools lack explicit input parameter type definitions in visible code, only high-level descriptions provided. No output schemas documented. Error handling and recovery guidance absent. Parameter descriptions exist but lack format constraints, ranges, and validation rules. No tool annotations (readOnlyHint, etc.). Composition is sound (each tool has one responsibility), but schema rigor falls short of production baseline.
Query Datadog APM traces for performance analysis and debugging
Search centralized logs in Datadog with trace correlation
Query infrastructure and application metrics from Datadog
Query Sentry for unresolved issues and errors for a specific service
Search system logs across multiple services with time range filtering and pattern matching
Input schemas lack explicit type definitions and constraints. Parameters documented in descriptions only; no JSON Schema with minLength, maxLength, pattern, enum, or range constraints visible in source code.
No output schemas documented. LLMs cannot infer response structure, field types, or pagination behavior. Agents must guess what fields to extract.
No error handling guidance. Tools lack recovery hints (e.g., 'If Datadog not initialized, set DD_ENABLED=true'). Errors will not guide LLM to next action.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | <=2025-11-25 | v2 |
Parameter descriptions lack format constraints and ranges. E.g., 'time_range' accepts '2 to 4pm' or '14:00 to 16:00' but no regex/format specified; 'limit' unbounded (no min/max).
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). All tools are read-only but this is not declared in schema, forcing LLMs to infer safety.