shiphrahx/ai-for-engineering-leaders/candidate-evaluation-summary

Claude Code and agent skill by shiphrahx/ai-for-engineering-leaders

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Produces a fair, evidence-based candidate evaluation summary — snapshot, strengths and concerns each tied to a specific interview observation, level assessment, team fit, and a hire/no-hire recommendation with confidence and conditions — useful to someone who wasn't in the interviews and fit for the candidate's file. Use when the user says "write up the evaluation", "candidate summary", "synthesize the interview feedback", or pastes feedback from multiple interviewers. Use this to WRITE the evaluation document — not to facilitate the decision meeting (interview-debrief-facilitation), nor to justify comp to HR (offer-justification).

Install

npx vskill@latest install shiphrahx/ai-for-engineering-leaders/candidate-evaluation-summary

SKILL.md

Candidate Evaluation Summary

Synthesize all interview feedback into one coherent, evidence-based evaluation — every strength and concern tied to an observation — that supports the hiring decision and stands up in the candidate's file.

Inputs to gather

Gather these before writing. If any are missing, ask for them in a single batched question — never invent feedback, ratings, or interview observations. Mark anything genuinely unavailable as Unknown in the output.

  • Candidate — name
  • Role and target level — what they applied for
  • Interview feedback — all interviewer notes (can be unstructured) and ratings, by stage
  • Interview dates — if available

Steps

  1. Read all feedback before writing. Tie every claim to a specific interview observation; if a claim has no supporting evidence, drop it or mark the evidence Unknown — do not infer.
  2. Analyze across interviewers: identify consistent signals (2+ interviewers agree), weight evidence by interview type for the skill in question (technical interview > cultural for technical depth), surface contradictions and state which signal is stronger and why, and distinguish "didn't demonstrate" from "demonstrated a weakness."
  3. Write the candidate snapshot: name, role, interview dates (or Unknown), and overall recommendation.
  4. Extract 3-5 strengths, each with the evidence and the interviewer/stage it came from, and a signal-strength label where the feedback supports it (e.g. "Strong signal"). Generic praise needs a concrete example ("explained a complex system clearly without jargon"), not an adjective.
  5. Extract 2-3 concerns, each with evidence and a severity assessment — minor (coachable), moderate (needs a plan), major (blocker) — and distinguish a real skill gap from a bad-interview artifact (nerves, a poor question, an off day).
  6. Write the level assessment: does the candidate meet the bar for the target level? Map evidence to level expectations; note absence of higher-level signals where relevant and whether that's expected.
  7. Write team fit: how they'd complement the existing team and which gaps they fill — concretely, not as personality praise.
  8. Write the recommendation: hire/no-hire, confidence level, and any conditions. For a borderline call, be explicit about what tips the balance ("despite concern X, hire because Y"). If one interviewer is a strong no, dig into the specifics — one strong signal may outweigh several weak positives.
  9. Adapt to context: apply the same evidence rigor to internal candidates — no softer bar because you know them; keep the document factual and professional (HR, legal, or the candidate may read it in some jurisdictions); for a rejected candidate who might reapply, note specific growth areas and what to look for next time.
  10. Assemble the output in the format below.

Output format

**Candidate Evaluation: [Name] — [Role]**
**Recommendation: [Hire / No Hire] ([Confidence: High/Medium/Low])**
**Strengths**
- **[Strength] ([signal strength]):** [Evidence, with interviewer/stage and quote if available.]
**Concerns**
- **[Concern] ([Minor/Moderate/Major]):** [Evidence.] Severity: [why, coachable or not].
**Level Assessment**
[Does the candidate meet the target-level bar? Evidence mapped to level expectations.]
**Team Fit**
[How they complement the team; which gap they fill; collaboration style.]
**Conditions:** [None — proceed to offer, or specific conditions before proceeding.]

Boundaries

  • Never state a strength or concern without an interview observation behind it — unsupported claims are dropped or marked Unknown.
  • Never fabricate ratings, quotes, dates, or interviewer names.
  • Never apply a softer standard to internal or known candidates.
  • Never assess interviewer calibration or verify claimed experience yourself — evaluate only demonstrated ability in the interviews; the user knows which interviewers run harsh or lenient.
  • Never include protected-characteristic or personal commentary; this document may be read by HR, legal, or the candidate, and compliance varies by jurisdiction.

Chaining

  • After a hire recommendation, offer offer-justification to justify level and comp to HR/compensation review.
  • This summary is typically the input to the debrief's documented decision — if the decision meeting hasn't happened, run interview-debrief-facilitation first.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
73
Repository forks
5
7-day trend score
22

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/shiphrahx/ai-for-engineering-leaders/candidate-evaluation-summary/badge)](https://verified-skill.com/skills/shiphrahx/ai-for-engineering-leaders/candidate-evaluation-summary)

Something wrong with this skill or its page? Tell us.