Resume Parsing Server has significant definition quality gaps. Tool names lack clear action verbs (parse_resume, match_resume_to_job are acceptable but inconsistent with verb_noun pattern). Descriptions are present but generic (50-100 chars), lacking WHEN/WHY context. No parameter descriptions visible for tools with actual schemas. No output schemas documented. Error handling is absent from tool definitions. Only 2 tools (match_resume_to_job, batch_match_resumes, anonymize_resume, compare_resumes, get_multi_resume_insights, export_resume) have non-empty schemas, but even these lack parameter descriptions in the visible code.

• Add non-empty input schemas to all 7 tools. At minimum, parse_resume should document required S3 bucket/folder parameters; anonymize_resume should define anonymization_level enum with descriptions.
• Add parameter descriptions to all schema fields. For export_resume, document that format_type must be one of ['json', 'xml', 'csv', 'ats_json'] and ats_system is required when format_type='ats_json'.
• Document output schemas for each tool. E.g., parse_resume should return {success: bool, resumes: [{id, name, skills, experience, education}], error?: string}.
• Expand tool descriptions to 100-200 chars with WHEN/WHY context. E.g., 'Match a resume against a job description to calculate compatibility score (0-100). Use this after parsing resumes to rank candidates. Returns score, matched_skills, missing_skills, and recommendations.'
• Add error handling guidance to descriptions. E.g., 'Returns error if resume_id not found (user should call parse_resume first) or job_description is malformed (must include title and required_skills).'

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

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