RepairSafe — Home Repair Safety Assistant

AI201 Lab 4 Starter Repository

RepairSafe is a home repair Q&A tool with a safety classification layer. Before answering any question, it classifies the request into one of three safety tiers and adjusts its behavior accordingly.

---

Setup

1. Fork this repo and clone your fork locally

2. Create and activate a virtual environment:

   python -m venv .venv
   source .venv/bin/activate   # Mac/Linux
   # or: .venv\Scripts\activate  # Windows

3. Install dependencies: pip install -r requirements.txt

4. Copy .env.example to .env and add your Groq API key

5. Run the app: python app.py

---

What to Implement

Milestone	File	Function	Description
1	safety.py	classify_safety_tier()	Classify question into safe / caution / refuse
2	responder.py	generate_safe_response()	Generate tier-appropriate response
3	auditor.py	log_interaction()	Append interaction record to audit log

Complete each spec in specs/ before implementing the corresponding function.

---

Repository Structure

ai201-lab4-repairsafe-starter/
├── app.py              ← Gradio UI and pipeline orchestration (pre-built)
├── safety.py           ← Milestone 1: safety tier classifier
├── responder.py        ← Milestone 2: tier-aware response generator
├── auditor.py          ← Milestone 3: audit logger
├── config.py           ← constants (API key, model, log path, valid tiers)
├── logs/               ← audit.jsonl written here after Milestone 3
└── specs/
    ├── system-design.md    ← read this first
    ├── classifier-spec.md  ← Milestone 1 spec
    ├── responder-spec.md   ← Milestone 2 spec
    └── auditor-spec.md     ← Milestone 3 spec