name: interview-kit-builder\ndescription: \"Generate a complete structured interview kit for a role — 3-5 role-specific competencies, one behavioral (STAR-format) question per competency, 1-5 scoring rubric with explicit behavioral anchors at each level, per-panel scorecards, interviewer debrief template.\"\nversion: \"1.0.1\"\ncategory: analysis\nplatforms:
\nYou generate a complete structured interview kit. The 2026 evidence: structured rubric-based interviews improve hiring accuracy 34% (Journal of Applied Psychology) and 87% of employers report behavioral interviews as their primary assessment method (NACE 2026). The gap between \"we did interviews\" and \"we ran a structured loop\" predicts hire performance better than years of experience or credentials.
\nA structured interview means: same questions, same rubric, same panel composition, calibrated scoring. Anything else is unstructured chat with a candidate.
\nVerify:
\nRecovery:
\nExtract 3-5 competencies from the JD's must-haves. Examples by role:
\nSenior Backend Engineer:
\nSr. PM:
\nB2B AE:
\nEach competency must be observable — i.e., you can describe what \"good\" looks like via behavior, not credentials.
\nVALIDATION: ≤ 5 competencies. Each has a one-sentence behavioral definition.
\nOne behavioral question per competency. STAR = Situation, Task, Action, Result.
\nTemplate:
\n\n\n\"Tell me about a time when [specific challenging situation that maps to this competency]. What was the [stakes/constraint]? What did you do? What was the outcome — and what would you do differently?\"
\n
Examples:
\nSystem design at scale:
\n\n\n\"Tell me about the highest-traffic system you've designed or significantly refactored. What were the load characteristics, the SLOs, and the biggest design trade-off you made? Looking back, what would you change?\"
\n
Customer discovery (PM):
\n\n\n\"Walk me through a time when customer research changed your roadmap. How did you choose who to interview? What was the original hypothesis vs what you learned? What did you ship as a result?\"
\n
Forecast accuracy (AE):
\n\n\n\"Describe a quarter where your forecast was significantly off — either over or under. What information were you missing? What's your process now to catch that signal earlier?\"
\n
Per question, include 3-5 follow-up probes the interviewer should use to dig deeper if the candidate stays high-level.
\nVALIDATION: Every competency has exactly one primary question + ≥ 3 follow-up probes. Questions don't reference protected categories.
\nFor each question, define what each score level looks like — not just \"good\" / \"bad\" but the specific signals.
\nTemplate (system design example):
\n| Score | \nBehavioral Anchor | \n
|---|---|
| 5 | \nDrew the system from scratch, identified the bottleneck before being asked, explained the failure modes, proposed a measurable rollout plan, and connected design choices to business outcomes. | \n
| 4 | \nDrew the system cleanly, named at least one significant trade-off and articulated why. Some failure modes considered. | \n
| 3 | \nCould describe a system they worked on, but didn't independently surface trade-offs without prompting. | \n
| 2 | \nConfused major concepts (e.g., consistency vs availability, latency vs throughput). Couldn't sketch a clean design. | \n
| 1 | \nCould not engage with the design question; deferred to \"we used X service\" without depth. | \n
VALIDATION: Each score has a behavioral anchor, not \"exceeds expectations.\" Anchor describes observable evidence.
\nGenerate a scorecard per interviewer in the loop:
\n# Scorecard — {Role} — {Interview Name}\n\nCandidate: {name}\nInterviewer: {name}\nDate: {date}\n\n## Competencies Assessed\n\n- {Competency 1}: \\_\\_\\_/5 (one anchor sentence with specific evidence)\n- {Competency 2}: \\_\\_\\_/5\n\n## Notable Strengths (specific behaviors observed)\n\n-\n-\n\n## Notable Concerns (specific behaviors observed)\n\n-\n-\n\n## Reservations / Open Questions\n\n-\n\n## Recommendation\n\n- [ ] Strong hire\n- [ ] Hire\n- [ ] No hire\n- [ ] Strong no hire\n\n(Pick one. \"Lean hire\" / \"lean no hire\" forbidden — calibration shows these collapse to \"hire\" 90% of the time. Force commitment.)\nVALIDATION: Each interviewer's scorecard covers ≤ 3 competencies (avoid one interviewer scoring all 5 — accuracy degrades).
\nGenerate a calibration session script for the panel BEFORE interviews start:
\nVALIDATION: Calibration script is ≤ 1 page, takes 30-45 min to run.
\nPer-loop debrief template (post-loop, all interviewers + recruiter + hiring manager):
\n# Debrief — {Candidate} — {Role}\n\n## Round-by-round scores\n\n| Round | Interviewer | Competency | Score | Key evidence |\n| ----- | ----------- | ---------- | ----- | ------------ |\n| Phone | Recruiter | Comm | 4 | ... |\n| HM | {name} | Leadership | 4 | ... |\n\nAverage competency score: X.X / 5\nMin competency score: X / 5\n\n## Discussion (5-10 min)\n\n- Strongest signal:\n- Weakest signal:\n- Outliers (any score ≥ 1 point off the panel mean): {who/what}\n- Reservations that didn't show up in writing:\n\n## Decision\n\n- [ ] Offer — {level} — {comp band}\n- [ ] No offer — primary reason: {one sentence}\n- [ ] Hold — additional reference call / second technical / etc.\n\n## If offer: assigned ramp manager + first-30-day plan creator\nVALIDATION: Debrief produces a single decision in writing, attributable, with rationale.
\nGenerate platform-specific exports:
\nVALIDATION: Generated file imports without errors into the target platform's sandbox.
\nScore 1–5:
\nCommon gap: rubric anchors written as \"exceeds/meets/below\" rather than observable behaviors. Rewrite each anchor with specific evidence.
\nAppend to ~/.claude/skills/interview-kit-builder/LEARNINGS.md: