README.md

# Technical Interview Copilot — v0.2.0

Skills-only OpenAI plugin for role-aware technical interview coaching. Compatible portable manifest with optional legacy overlay; no remote MCP server, OAuth or external execution dependency. Public name and package identifier preserved for updating the existing listing.

## How it works
- New JD → `interview-map`; timed preparation → `prep-plan` or `rapid-review`.
- Mock interview → `mock-interview`, one question at a time.
- Coding, system design, SQL/data, incident and AI/ML → specialized practice skills.
- Resume stories → `experience-deep-dive` or `behavioral-interview` with truthful evidence.
- Company research → current official sources with uncertainty, not unverified generic interview loops.
- Post-answer review → `feedback-and-rubric`; assessment interpretation → `take-home-assessment`.
- Actual subagents, code execution and private document access only where the user's host makes them available and authorizes their use.

## Source and quality
`references/original-v0.1.0/` contains unmodified original skill and 4 reference documents. `UPGRADE_AUDIT.md` records changes. `benchmarks/cases.json` contains 24 prospective acceptance scenarios, not claimed as executed. Optional offline scripts are deterministic coaching helpers, not hiring predictions.

## Checks
`python3 scripts/validate_package.py` and `python3 -m unittest discover -s tests -v` from package root. ZIP integrity check separately. These do not prove installed plugin behavior.

## Support and publisher
Tekris Labs — https://tekrislabs.com/ ; contact https://tekrislabs.com/contact ; privacy https://tekrislabs.com/privacy ; terms https://tekrislabs.com/terms . All web URLs must be accessible before public submission.