菜单
Improve existing behavioral interview specs in this repository. Use when upgrading an interview JSON file, making generic examples more concrete, strengthening strong and weak answer patterns, deepening follow-up probes, improving evaluation rubrics, or aligning an existing interview with the project schema and quality bar.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Add or improve Markdown bold emphasis for key phrases in behavioral interview JSON text. Use when highlighting competency briefs, prompts, prompt variants, interviewer intent, evidence, decision criteria, trade-offs, operating mechanisms, red flags, follow-up probes, or rubric signals with **key phrase** markers while preserving valid project data.
Promote a specific behavioral interview on LinkedIn from a provided interview.json URL or path. Use when drafting LinkedIn copy that explains behavioral interviews, their education and interviewing value, the project, and includes both the concrete interview anchor link and the full catalog link.
Review behavioral interview specs in this repository for realism, balance, educative value, interviewer usability, and data quality. Use when asked to critique interviews, audit the catalog, identify gaps, score interview quality, or write/update REVIEW.md with findings and recommendations.
Review behavioral interview specs in this repository for realism, balance, educative value, interviewer usability, and data quality. Use when asked to critique interviews, audit the catalog, identify gaps, score interview quality, or write/update REVIEW.md with findings and recommendations.
Create new behavioral interview specs for the spec-driven-behavioral-interviews repository. Use when adding a new interview dataset, choosing category placement, scaffolding data/book or data/examples entries, filling the JSON schema, or preparing a new interview for the static explorer.
Add or improve concrete example stories, prompt variants, weak and strong answer variants, and follow-up angles for behavioral interview specs in this repository. Use when a user asks for more concrete examples inside each interview, more variants by role or level, or better example coverage.
| name | behavioral-interview-improver |
| description | Improve existing behavioral interview specs in this repository. Use when upgrading an interview JSON file, making generic examples more concrete, strengthening strong and weak answer patterns, deepening follow-up probes, improving evaluation rubrics, or aligning an existing interview with the project schema and quality bar. |
data/<group>/<id>/interview.json, its group manifest, AGENTS.md, and adjacent interviews in the same competency family.python3 _scripts/summarize_coverage.py
python3 _scripts/validate_interviews.py
data/. Preserve ids and manifest paths unless the user explicitly asks to rename.python3 build.py after source changes.Strengthen the spec until it answers these questions:
Prefer concrete operating mechanisms over adjectives. Replace "improved communication" with a forum, cadence, decision rights, escalation rule, dashboard, ownership model, or planning change. Replace "drove alignment" with who disagreed, what criteria resolved the disagreement, and what decision became easier afterward.
Keep example stories adaptable. They should be story shapes a candidate can map to their experience, not scripts to memorize.
Run the checks relevant to the edit:
python3 _scripts/validate_interviews.py
python3 _scripts/summarize_coverage.py
python3 build.py
python3 _scripts/smoke_static_site.py