Resume MCP has 3 tools with basic schemas and descriptions, but significant gaps prevent production use. All tools have descriptions (10-50 chars, below the 50-200 char baseline for LLM optimization) and input schemas with type definitions. However, parameter descriptions are minimal or absent, output schemas are undocumented, and error handling is not visible. The tool names are action-oriented (analyze_, match_, rank_) but descriptions lack context on when to use each tool vs. alternatives. No evidence of pagination, result limits, or recovery guidance in error cases. Source code shows FastAPI setup but tool implementation details (resume_analyzer.py) are not provided, making it impossible to verify actual behavior, error handling, or output structure.

• Expand tool descriptions to 80-150 chars. Example: 'analyze_resume: Extract structured information from a resume (skills, experience, education, certifications). Use this first to understand candidate qualifications before matching to jobs. Returns JSON with extracted fields and confidence scores.'
• Add parameter descriptions for all inputs. Example: 'resume_text: Full resume text (plain text or markdown, 100-50000 chars). Can be extracted from PDF via /upload-resume/ endpoint.'
• Document output schemas explicitly. Define what each tool returns: field names, types, and meaning. Example: 'analyze_resume returns {skills: [string], experience: [{title, company, duration}], education: [{degree, school}], confidence: 0-1}'.
• Add error handling guidance. Example: 'If resume_text is empty or unreadable, returns {error: "Invalid resume format", suggestion: "Upload a PDF or paste plain text"}. Retryable: yes.'
• Implement result limits and pagination for rank_candidates. Cap at 50 results, return total_count and next_cursor. Document in description: 'Returns top 50 ranked candidates. Use next_cursor to fetch additional results.'

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry