Claude Skill

Resume Screener

Scores a resume against role criteria with a structured rubric — not vibes.

Download skill (.zip)Or download whole pack

What it does

Given a role spec (must-haves, nice-to-haves, deal-breakers) and a resume, produces a structured screen: pass/maybe/no with cited evidence from the resume. Calls out gaps explicitly. Flags promotion velocity, gaps in employment, and credentials that don't match level claims.

When to use

✓You have a stack of resumes and need consistent first-pass screening
✓A hiring manager wants a second opinion before reaching out
✓Calibrating your own screening — does the AI agree with your gut?

When not to use

✗Final hiring decision. This is a screen, not a hire.
✗Resumes for very specialized roles (deep research, exec) where context matters more than pattern-matching
✗You don't have a clear role spec yet

Install

Download the .zip, then unzip into your Claude skills folder.

mkdir -p ~/.claude/skills
unzip ~/Downloads/resume-screener.zip -d ~/.claude/skills/

# Restart Claude Code session.
# Skill is now available — Claude will use it when relevant.

SKILL.md

---
name: resume-screener
description: Use when screening one or more resumes against a role specification. Triggers on "screen this resume", "score this resume", "evaluate this candidate", or pasted resumes alongside a JD.
---

# Resume Screener

Screen resumes consistently against a role spec. Output is structured evidence, not opinion. The goal is to filter the funnel, not to make the final hire.

## Required inputs

1. **Role spec** with explicit:
   - Must-haves (gating)
   - Nice-to-haves (signal)
   - Deal-breakers (auto-no)
   - Seniority expectation
2. **One or more resumes** (or LinkedIn dumps)

If the role spec is vague, ask before screening. Do not invent criteria.

## Screening framework

For each resume, produce this structure:

### Verdict
**PASS / MAYBE / NO** + 1-sentence rationale.

### Must-haves
For each must-have:
- ✅ Met — cite the specific role/project/year
- ⚠️ Partial — explain the gap (e.g. "5y in industry but only 2y in payments domain")
- ❌ Missing — flag explicitly

### Nice-to-haves
List which are present, with evidence.

### Seniority calibration
- Years of total experience
- Promotion velocity (e.g. "IC → Senior in 2y is fast; → Staff in 3y is fast")
- Scope progression (team size, system size, scope of ownership)
- **Mismatch flag** if the title doesn't match the work described

### Concerns
Explicit list:
- Employment gaps > 6 months (note them, don't penalize without context)
- Tenure pattern (e.g. "4 jobs in 3 years")
- Title inflation (e.g. "Senior" with 2 YOE at a tiny company)
- Vague impact claims ("led initiatives", "drove growth" with no numbers)

### Strengths beyond the spec
Anything genuinely interesting that wasn't asked for but raises priority.

### Suggested screening questions
2-3 questions a recruiter should ask on the phone screen specifically based on this resume.

## Bias controls

- **Never** infer or comment on age, gender, ethnicity, or nationality.
- Don't penalize non-traditional career paths without specific role-relevant reason.
- Don't reward prestigious schools/companies if the must-haves aren't met.
- Don't penalize lack of prestigious schools/companies if the must-haves are met.

## Tone

Evidence-first. Quote specific lines from the resume when making a claim. If a claim is unsupported ("led 5x growth" with no context), flag it for the recruiter to probe — don't accept or reject on it.

## Output format

Use the structure above. For batches, end with a ranked list (best to worst pass).

Example prompts

Once installed, try these prompts in Claude:

Screen this resume against our Senior Backend Engineer role: must-have Go + 5y experience + payments domain. Nice: Kafka, Postgres at scale. [paste resume]
Score these 5 resumes against our PM role spec. Rank pass/maybe/no with reasoning.