resume-score

Score a resume against a job description. Be candid — a low score is useful information, not something to soften.

Invoke the resume-coach agent to perform the scoring analysis.

Workflow

1. Collect inputs

Resume — required. Accept:

• A file path to an existing document in {user_dir}
• An artifact name from CareerNavigator/artifacts-index.json (look up by filename)
• Content pasted directly into the conversation

Job description — required. If not provided:

"Paste the job description to score against."

If the user only provided a resume with no JD, and there is an active role in context (e.g., from a recent search-jobs result or tracker entry), ask:

"Which job description should I score this against?"

2. Invoke resume-coach

Pass the resume text and JD to resume-coach for analysis. The agent applies its ATS compatibility, narrative strength, and strategic fit scoring rubrics and returns scores per dimension.

3. Present the score

**Resume Score** — {Resume filename} vs. {Company/Role or "provided JD"}

Overall: {n}/100

ATS Compatibility    {n}/40
  {keyword match count, formatting flags, one line each}

Narrative Strength  {n}/35
  {2–3 sentences of candid assessment: achievement quality, summary relevance, seniority signal}

Strategic Fit       {n}/25
  {1–2 sentences: role alignment and differentiator visibility}

---
Top gaps to address
1. {Missing keyword or structural issue} — {why it matters}
2. ...
3. ...

Cap the gaps list at 5. Order by impact on ATS pass rate first, narrative second.

4. Offer next steps

After the score, offer:

"Want me to tailor a new version of this resume for the role? Run /career-navigator:tailor-resume or just say 'tailor it'."

Do not automatically trigger tailoring — scoring and tailoring are distinct operations. Let the user decide.