cv-cover-optimizer

name	cv-cover-optimizer
version	0.2.0
description	Optimiza CV y carta de presentacion para ATS, rol objetivo y voz de marca con lint offline de keywords, secciones, longitud, contacto y tono.
owner	JM Labs
triggers	["cv-cover-optimizer","optimizar-cv","ats-cv"]
allowed-tools	["Read","Write","Edit","Bash"]

Cv Cover Optimizer

Purpose

Use this skill to improve a CV, resume, cover letter, or application packet against a target role. The skill produces ATS-aware edits, coverage gaps, brand-voice guardrails, and a deterministic validation result without inventing credentials.

Inputs Expected

• Candidate CV or cover text.
• Target role, company, job description, or required keywords.
• Language preference and brand voice constraints.
• Permission boundary for editing files or returning proposed copy only.

Outputs Expected

• Optimized CV or cover letter sections.
• ATS keyword coverage summary.
• Missing section/contact/length issues.
• Brand voice risks, including stacked trademarks or hustle framing.
• Validation command evidence from scripts/ats_lint.py when a machine-readable packet is provided.

Procedure

Discover

Identify the target role, source document type (cv or cover), required keywords, language, and whether the user wants direct edits or proposed text.

Analyze

Compare source text against assets/ats-keyword-policy.json, assets/document-section-policy.json, assets/brand-voice-policy.json, and assets/privacy-policy.json.

Execute

Rewrite only the requested sections. Preserve facts, seniority, dates, employers, credentials, and contact details. Do not invent metrics; mark suggested metrics as placeholders.

Validate

Run bash skills/cv-cover-optimizer/scripts/check.sh for fixture coverage. For a specific JSON packet, run:

python3 skills/cv-cover-optimizer/scripts/ats_lint.py --input <packet.json>

Assets

• assets/ats-keyword-policy.json
• assets/document-section-policy.json
• assets/brand-voice-policy.json
• assets/privacy-policy.json
• assets/output-contract.json

Quality Criteria

• Recommendations are grounded in source text and target role evidence.
• ATS gaps identify missing keywords without keyword stuffing.
• CV output includes recognizable experience, education, skills, or achievement sections.
• Cover output stays concise and avoids generic flattery.
• Contact/PII handling is explicit and privacy-safe.
• Claims about candidate experience are not invented.

Edge Cases

• No job description: produce a blocked or partial optimization and ask for role evidence.
• Minimal CV: focus on structure and missing data rather than fabricating content.
• Conflicting brand tone: preserve user preference and mark unresolved conflicts.
• Sensitive personal data: avoid exposing direct contact details in examples or fixtures.
• Cover letter over length: prioritize evidence-backed value and remove repetition.

Scripts

scripts/ats_lint.py --input <json> lints CV/cover packets for ATS keyword coverage, section presence, length, contact, stacked trademarks, and hustle language. scripts/check.sh runs deterministic valid and invalid fixtures offline.

Related Skills

• cv-transformer
• brand-voice
• gratitud-post-proceso

Evidence Requirements

• Cite source document, job description, or user-provided constraints for every substantive recommendation.
• Mark inferred gaps as assumptions when the target role evidence is incomplete.

Update-Safety Notes

• Default to proposed copy unless file-edit authority is explicit.
• Never overwrite the source CV or cover letter without explicit confirmation.
• Keep placeholders visibly marked when a metric or detail needs user confirmation.