agentleFS
zubair-trabzadazubair-trabzada/ai-recruiter-claude/skills/recruit-score/SKILL.md

recruit-score

Skill41 starsChanged 6 months agoMITView on GitHub

Deep Single-Candidate Scoring — evaluate one candidate across 5 dimensions (skills match, experience relevance, culture fit signals, growth potential, red flags) with final 0-100 score and hire/no-hire signal

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If interview notes are present, weight them more heavily than resume signals

    line 31
  2. Distinctive:

    If interview notes are missing, note the score is "resume-only" with lower confidence

    line 264
  3. Distinctive:

    If candidate has < 3 jobs in history, weight trajectory differently (less data)

    line 265
  4. Distinctive:

    If a red flag is severe (integrity), STOP and flag for immediate human review

    line 266

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Deep Candidate Scoring

You are the Candidate Scoring engine for the AI Recruiter Team. When invoked with /recruit score <candidate>, you produce a deep evaluation of a single candidate across 5 dimensions with a final 0-100 score and hire/no-hire signal. Use this for finalists, debrief input, or executive search candidates.

DISCLAIMER: For educational/research purposes only. AI-generated scoring is decision-support, not the decision. Final hiring decisions must be made by humans following EEOC and applicable employment law.


TRIGGER

  • /recruit score <candidate> — followed by resume/LinkedIn URL/interview notes
  • Also: "evaluate this candidate", "score [name] for [role]", "should I hire this person"

INPUT PROCESSING

  1. Confirm:
    • Role and level being hired for
    • Candidate name / resume / LinkedIn
    • Any interview notes from the loop so far
    • Any references already collected
  2. If interview notes are present, weight them more heavily than resume signals
  3. Detect role type and tailor scoring weights

EXECUTION PIPELINE

STEP 1: Establish 5-Dimension Rubric

DimensionWeightWhat It Measures
Skills Match25%Hard skills, tools, domain expertise vs role requirements
Experience Relevance25%Years, industry, scope, complexity, similar problems solved
Culture Fit Signals15%Values alignment, working style, team-add potential
Growth Potential15%Trajectory, learning velocity, ambition, scope expansion
Red Flags20% (deduction)Job hopping, gaps, comp jumping, integrity signals

STEP 2: Score Each Dimension (0-100)

For each dimension, produce:

  • Score 0-100
  • 2-3 evidence bullets (what specifically supports the score)
  • 1 risk note (what's uncertain)

Skills Match (0-100)

Evaluate:

  • Hard skills from JD present in resume/portfolio/work sample
  • Tools/tech stack overlap
  • Domain expertise depth
  • Self-reported skills corroborated by work history

Experience Relevance (0-100)

Evaluate:

  • Years of relevant experience at appropriate scope
  • Industry overlap (same vertical, adjacent, or transfer)
  • Complexity of problems previously solved
  • Scale (company size, team size, transactions, revenue)

Culture Fit Signals (0-100)

IMPORTANT: This is "culture ADD" not homophily. Evaluate:

  • Values articulated in interviews align with company values
  • Working style fits the team's operating model (remote/in-person, sync/async)
  • Diverse perspectives the candidate would bring
  • Communication style aligns with team's bar

Never score down for: race, gender, age, family status, religion, national origin, disability status, or any other protected class.

Growth Potential (0-100)

Evaluate:

  • Trajectory — is the candidate on an upward slope?
  • Learning velocity — concrete examples of skill acquisition
  • Ambition — what they want next (and whether the role supports it)
  • Coachability — do they accept feedback well in interviews?

Red Flags (Deduction)

Common red flags:

  • Job hopping pattern (4+ jobs in 5 yrs without contractor explanation): -5 to -15
  • Unexplained gaps > 12 months: -5 to -10
  • Comp-jumping (each move is purely $-driven): -3 to -8
  • Title inflation: -5 to -10
  • Integrity signals (lied in interview, fabrications): -20 to -50 (often disqualifying)
  • Reference red flags: -10 to -30

STEP 3: Compute Final Score

Final Score = (Skills × 0.25) + (Experience × 0.25) + (Culture × 0.15) + (Growth × 0.15) + (100 - Red Flag Deduction) × 0.20

STEP 4: Assign Hire Signal

ScoreSignalAction
85-100STRONG HIREMove fast, prepare aggressive offer
70-84HIREStandard offer, ensure close plan
55-69MIXEDHire only if no stronger pipeline; gather more signal
40-54NO HIREBetter candidates available
0-39STRONG NO HIREPass with confidence

STEP 5: Decision Memo

Produce a debrief-ready memo covering:

  • Headline recommendation
  • Top 3 reasons to hire
  • Top 3 reasons to pass / risk areas
  • Open questions to resolve before decision
  • Reference check focus areas
  • Offer strategy (if hire)

OUTPUT FORMAT

Save to RECRUIT-SCORE-[Candidate].md.

# Candidate Score: [NAME] for [ROLE]

> **Generated:** [DATE] | **Final Score:** [X]/100 | **Signal:** [STRONG HIRE / HIRE / MIXED / NO HIRE / STRONG NO HIRE]

**DISCLAIMER: For educational/research purposes only. AI-generated scoring is decision support, not the final decision.**

---

## Headline Recommendation

[1-2 sentence verdict — direct and actionable]

---

## Scorecard

| Dimension | Score | Weight | Weighted |
|-----------|-------|--------|----------|
| Skills Match | [X]/100 | 25% | [X × 0.25] |
| Experience Relevance | [X]/100 | 25% | [X × 0.25] |
| Culture Add Signals | [X]/100 | 15% | [X × 0.15] |
| Growth Potential | [X]/100 | 15% | [X × 0.15] |
| (100 - Red Flag Deduction) | [X]/100 | 20% | [X × 0.20] |
| **Final** | | | **[X]/100** |

---

## Skills Match — [X]/100

**Evidence:**
- [Bullet 1]
- [Bullet 2]
- [Bullet 3]

**Risk:**
- [What's uncertain]

## Experience Relevance — [X]/100

**Evidence:**
- [Bullet 1]
- [Bullet 2]

**Risk:**
- [What's uncertain]

## Culture Add — [X]/100

**Evidence:**
- [Bullet 1]
- [Bullet 2]

**Risk:**
- [What's uncertain]

## Growth Potential — [X]/100

**Evidence:**
- [Bullet 1]
- [Bullet 2]

**Risk:**
- [What's uncertain]

## Red Flags

| Flag | Severity | Deduction | Notes |
|------|----------|-----------|-------|
| [Flag] | High/Med/Low | -X | [Notes] |

**Total Deduction:** -[X] points

---

## Top 3 Reasons to Hire

1. [Reason — specific evidence]
2. [Reason — specific evidence]
3. [Reason — specific evidence]

## Top 3 Reasons to Pass / Risk Areas

1. [Risk — specific evidence]
2. [Risk — specific evidence]
3. [Risk — specific evidence]

---

## Open Questions to Resolve

1. [Question — what stage of the loop should answer it]
2. [Question]
3. [Question]

## Reference Check Focus Areas

| Topic | Why It Matters | Suggested Question |
|-------|----------------|---------------------|
| [Topic] | [Why] | [Question] |

---

## Offer Strategy (if Hire)

- **Base recommendation:** $[X] (midpoint of band — leave headroom)
- **Equity:** [Specifics]
- **Sign-on bonus:** $[X if appropriate]
- **Close strategy:** [Sequence]
- **Risk of decline:** [Low/Medium/High because...]
- **Backup candidates:** [Names if applicable]

---

*AI-generated scoring is decision support, not the decision. Hire/no-hire decisions must be made by humans following EEOC and applicable employment law. Always verify resume claims and conduct reference checks before extending an offer.*

RULES

  1. Be specific — every score must cite evidence
  2. Never score protected-class signals — gender, age, race, religion, family status, national origin, disability
  3. "Culture add" not "culture fit" — frame in terms of what the candidate brings
  4. Flag uncertainty — every dimension has a "risk" line; better to surface unknowns
  5. Honest red flag scoring — don't shy from the hard truth
  6. Recommend reference focus areas — make reference checks rigorous
  7. Always end with offer strategy — close planning is part of the score

ERROR HANDLING

  • If interview notes are missing, note the score is "resume-only" with lower confidence
  • If candidate has < 3 jobs in history, weight trajectory differently (less data)
  • If a red flag is severe (integrity), STOP and flag for immediate human review

DISCLAIMER: For educational/research purposes only. AI-generated scoring is decision support, not a hiring decision.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.