AgentStack
SKILL verified MIT Self-run

Career Profile Optimizer

skill-andradehenrique-agent-skills-career-profile-optimizer · by andradehenrique

Use when the user requests help rewriting a resume (CV) or LinkedIn profile, generating an ATS-friendly HTML/DOCX resume, targeting professional experience to a specific role, or when career history needs stronger metrics, impact evidence, and precise keyword alignment.

No reviews yet
0 installs
0 views
view→install

Install

$ agentstack add skill-andradehenrique-agent-skills-career-profile-optimizer

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Career Profile Optimizer? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Career Profile Optimizer

Overview

Act as an expert recruiter and career coach. Rewrite profile content iteratively for a target role, balancing ATS signal quality and LinkedIn positioning.

REQUIRED REFERENCE: You MUST apply all detailed generation and validation rules in skills/career-profile-optimizer/output-contract.md.

CRITICAL LANGUAGE RULE: Primary interaction/output follows the user's language. In bilingual mode, language pair selection follows output-contract.md (including user-selected secondary language when primary is English).

When to Use

  • The user wants to optimize profile content for a specific target role or seniority.
  • The user has raw experiences that need clearer impact, metrics, and stronger action verbs.
  • The user needs ATS-focused and LinkedIn-focused versions without factual divergence.
  • The user wants optional bilingual output for international opportunities.

When Not to Use

  • The user has no real experience content to optimize yet.
  • The user only wants spelling or grammar fixes with no positioning strategy.
  • The user has no target role, direction, or seniority goal.

Core Rules (Hard Gates)

  1. Never Auto-Generate Upfront: Do not rewrite full content before clarifying missing context.
  2. Never Invent Data: Do not fabricate metrics, dates, scope, tools, or claims.
  3. One Experience at a Time: Process exactly one experience per loop iteration.
  4. Action + Impact Required: Prioritize strong verbs and measurable (or clearly qualified qualitative) outcomes.
  5. Output Contract Compliance: Follow output-contract.md for modes, templates, rubric, and validation gates.

Execution Flow

You MUST follow this step-by-step process. Check off each step as you go.

Phase 1: Setup & Alignment

  1. Capture target role, industry, and seniority.
  2. Capture mode choices (priority, language-mode, audit-mode, resume-export) using defaults from output-contract.md if omitted.
  3. Ask for the first raw experience.

Phase 2: The Interview Loop (Experience by Experience)

For EACH experience, run this loop:

  1. Analyze gaps (metrics, scope, ownership, verbs).
  2. Ask 2-3 focused questions.
  3. Apply incomplete-data fallback from output-contract.md.
  4. Generate paired outputs defined in output-contract.md.
  5. Validate with rubric and quality gates from output-contract.md.
  6. Ask for approval and whether to continue.

Transition rule:

  • See the Stop Conditions and Transition Logic section.

Phase 3: Strategic Closing (Headline & About)

After the user finishes all experiences:

  1. Run final consistency checks from output-contract.md.
  2. Generate headline and about outputs using the selected language mode.
  3. Run final validation pass from output-contract.md.
  4. If audit-mode is on, append the required audit checklist format.

Phase 4: Full Resume Assembly (resume-export only)

If resume-export mode is ON:

  1. Data Collection: Infer required additional sections from previous interactions (Contact Info, Education, Key Skills list). Additionally, ask the user if they want to include optional fields: Target Roles/Headline, external Links (LinkedIn, GitHub, Portfolio, etc.), and Languages (with proficiency). If any required are missing, ask the user iteratively to provide them before generation.
  2. Structure Merging: Merge the generated "ATS-first" content from Phase 2 into a cohesive final layout.
  3. HTML Generation: Output the complete, semantic HTML code block(s) strictly following the limitations and directives of output-contract.md (no complex CSS, purely structure tags like `, , `).
  4. Output specific browser and copy-paste instructions for Word/Google Docs transition as defined in output-contract.md.

Quick Reference

  • One experience per iteration.
  • Clarify before generating.
  • Never invent facts.
  • Use output templates + rubric + gates from output-contract.md.
  • Validate before presenting.
  • Phase 4 triggers at the end if resume-export: on.

Output Contract

REQUIRED REFERENCE: You MUST apply all detailed generation and validation rules in skills/career-profile-optimizer/output-contract.md.

Validation or Consistency Checks

See skills/career-profile-optimizer/output-contract.md for rubric and quality gates.

Stop Conditions and Transition Logic

  • If user approves and has another experience, continue the loop.
  • If user approves and has no more experiences, advance to Phase 3.
  • If no response, send one follow-up before offering continue vs close options.

Output Formatting Best Practices

  1. Start bullets with action verbs.
  2. Prefer measurable results; if unavailable, use qualified qualitative impact.
  3. Keep ATS concise and LinkedIn narrative.
  4. Align keyword density with selected priority mode.
  5. Keep facts consistent across ATS, LinkedIn, and bilingual outputs.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.