360 Secure SkillHub
360 安全技能中心

Interview

面向求职面试的准备系统,提供公司调研、STAR 故事库构建、模拟面试与反馈、薪资谈判策略及跟进邮件起草,数据仅本地保存于 memory/interview/。包内多个参考文档与脚本缺失,部分声明的能力无法实际执行。

agistack
agistack
热度 93.06
最近更新 2026-05-11 17:29:28
分类:知识依赖:代码执行 / 读写文件应用场景:教学辅导 / 研究 / 邮件起草 / 报告撰写 ...

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add interview
Metadata
name
interview
description
Interview preparation system with company research, story building, and mock interview practice. Use when user mentions job interviews, interview prep, behavioral questions, salary negotiation, or follow-up messages. Researches companies, builds story libraries, runs mock interviews, prepares salary strategies, and drafts follow-ups. NEVER guarantees job offers.

Interview

Interview mastery system. Preparation that wins offers.

Critical Privacy & Safety

Data Storage (CRITICAL)

  • All interview data stored locally only: memory/interview/
  • No external job platforms connected
  • No application tracking systems integrated
  • No sharing of interview content
  • User controls all data retention and deletion

Safety Boundaries

  • ✅ Research companies and roles
  • ✅ Build story libraries from experience
  • ✅ Run mock interviews with feedback
  • ✅ Prepare salary negotiation strategies
  • ❌ NEVER guarantee job offers
  • ❌ NEVER provide false information
  • ❌ NEVER replace genuine preparation

Data Structure

Interview data stored locally:

  • memory/interview/research.json - Company research briefs
  • memory/interview/stories.json - Story library
  • memory/interview/practice.json - Mock interview records
  • memory/interview/salary.json - Salary research and strategies
  • memory/interview/feedback.json - Post-interview notes

Core Workflows

Research Company

User: "Research Acme Corp for my interview Friday"
→ Use scripts/research_company.py --company "Acme Corp" --role "Product Manager"
→ Generate comprehensive research brief with talking points

Build Story

User: "Help me build a story about the project failure"
→ Use scripts/build_story.py --situation "project-failure" --lesson "learned"
→ Structure STAR format story with specific details

Mock Interview

User: "Run a mock interview for PM role"
→ Use scripts/mock_interview.py --role "Product Manager" --level senior
→ Ask realistic questions, provide honest feedback

Prepare Salary

User: "How should I handle the salary question?"
→ Use scripts/prep_salary.py --role "Product Manager" --location "SF"
→ Research market data, prepare negotiation strategy

Draft Follow-up

User: "Draft thank you email for today's interview"
→ Use scripts/draft_followup.py --interview "INT-123" --tone professional
→ Generate specific, memorable follow-up message

Module Reference

Scripts Reference

Script Purpose
research_company.py Generate company research brief
build_story.py Build STAR format stories
mock_interview.py Run practice interview
prep_salary.py Prepare salary strategy
draft_followup.py Draft follow-up messages
analyze_role.py Analyze job description
identify_gaps.py Identify experience gaps
log_feedback.py Log post-interview feedback