chengzicv/chengzicv-career-skills / chengzicv-interview-mock
chengzicv-interview-mock
Verified通过五题模拟面试生成企业初筛诊断;在用户需要 HR 面或综合面训练时使用。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 9.8 KB |
| agents/openai.yaml | config | 271 B |
| assets/mock-report-template.html | other | 12.0 KB |
| onboarding/core-introduction.md | markdown | 3.8 KB |
| onboarding/full-methodology.md | markdown | 1.8 KB |
| prompts/diagnostician.md | markdown | 1.9 KB |
| prompts/evidence-extractor.md | markdown | 972 B |
| prompts/interviewer.md | markdown | 3.1 KB |
| references/report-template.md | markdown | 851 B |
| references/runtime-contract.md | markdown | 1.8 KB |
| references/safety-and-claims.md | markdown | 683 B |
| runtime/input-contract.schema.json | config | 1.7 KB |
| schemas/evidence-pack.schema.json | config | 2.1 KB |
| schemas/interview-turn.schema.json | config | 525 B |
| schemas/report.schema.json | config | 4.3 KB |
| scorecards/dimensions.json | config | 1.5 KB |
| scorecards/scene-priorities.json | config | 1.0 KB |
| scripts/schema_subset.py | script | 4.1 KB |
| scripts/validate_evidence_pack.py | script | 2.6 KB |
| scripts/validate_report.py | script | 2.7 KB |
| strategy.json | config | 337 B |
references/runtime-contract.md
View on GitHub# 产品 Runtime 集成契约
## 一句话职责
平台保存状态,Skill 生成行为;Evidence Pack 在面试完成后一次性产生,只服务报告。
## 平台必须持有
- `session_id`
- 简历、JD、岗位与公司信息
- 完整原始 `messages`,每条有稳定 `message_id`
- `current_question_id`
- `followup_count`
- `completed_question_ids`
- `onboarding_version_seen`
- `skill_version`
- 面试、提证、诊断每次调用的 run 记录
## 推荐调用
### 面试调用
每轮传入材料、session state、完整 transcript(五题 MVP 可全量传入)和当前回答。模型返回 `interview-turn`。平台保存模型消息并更新题号/追问计数。
不要在此阶段生成 Evidence Pack。
### 提证调用
五题完成后,使用完整 transcript 发起一次独立调用,生成 Evidence Pack。运行逐字校验;失败时带校验错误重试,最多两次。
### 诊断调用
使用一次新调用,传入材料、完整 transcript、已校验 Evidence Pack 和评分规则。生成报告后运行 Schema 与 evidence_ref 校验。
## 恢复
恢复只依赖平台的 messages 和 session state。若面试未完成,从未完成题继续;若面试完成而报告缺失,重新执行提证和诊断。
## 失败与幂等
- 所有 run 带幂等键,重试不重复推进题号。
- 原始消息写入成功后才调用模型。
- 模型响应保存成功后再向前端确认完成。
- Evidence Pack 校验持续失败时,诊断回退为完整 transcript,并在报告中不展示无法核验的直接引用。
- 诊断失败不影响已保存的面试历史。
## 不要做
- 不依赖模型厂商 thread 作为唯一事实源。
- 不让 Skill 声称自己保存了聊天。
- 不用自由摘要替代原始 transcript。
- 不把 Evidence Pack 当作暂停恢复状态。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |