docs: add CLAUDE.md with codebase overview for AI assistants

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GQkuCo6PEN2NYUkJiCP5jG

Author: claude

CLAUDE.md

@@ -0,0 +1,60 @@
+# CLAUDE.md
+
+## Project Overview
+
+This is the **Maximum PC Builds Archive** — an archive of PC builds featured in Maximum PC magazine. Content is stored as Markdown files organized by year/month, with a companion website on the `gh-pages` branch. A Python scraper automatically updates part prices daily via GitHub Actions.
+
+## Repository Structure
+
+```
+├── 2018/, 2020/, 2021/   # Build archives by year → month → build markdown files
+├── docs/                  # Documentation assets (images, etc.)
+├── .github/
+│   ├── workflows/         # GitHub Actions (gitleaks, update-prices)
+│   ├── agents/            # GitHub Copilot agent configs
+│   └── ISSUE_TEMPLATE/    # Issue templates
+├── scraper.py             # PCPartPicker price scraper (Python 3.12)
+├── requirements.txt       # Python deps: beautifulsoup4, requests, lxml
+├── README.md              # Main docs with table of contents linking all builds
+└── CONTRIBUTING.md        # Contribution guidelines
+```
+
+## Build File Conventions
+
+- Each build is a Markdown file named by category: `AMD Budget.md`, `Intel Turbo.md`, etc.
+- Pre-2020 builds: Budget, Mid-Range, Turbo categories only.
+- 2020+: Separate AMD and Intel variants for each tier.
+- Files contain PCPartPicker list links and component tables with prices.
+
+## Development
+
+### Price Scraper
+
+```bash
+pip install -r requirements.txt
+python scraper.py          # Updates prices in all build markdown files
+DEBUG=1 python scraper.py  # Verbose logging
+```
+
+The scraper (`scraper.py`) extracts PCPartPicker URLs from markdown files, scrapes current prices, and updates the files in-place. It runs daily at 2 AM UTC via `.github/workflows/update-prices.yml`.
+
+### CI/CD
+
+- **update-prices.yml** — Daily automated price scraping, commits changes directly to main.
+- **gitleaks.yml** — Secret scanning on pushes/PRs.
+- **GitHub Pages** — The `gh-pages` branch serves the website.
+
+## Commit Message Convention
+
+Format: `type(scope): description`
+
+Examples from history:
+- `feat(pip): bump lxml to 6.1.0`
+- `fix(github-actions/workflows): update dependencies of gitleaks action to v3`
+
+## Key Guidelines
+
+- Never commit `.env` files or secrets.
+- When adding a new magazine issue: create markdown files in the appropriate `year/month/` folder, then update the README table of contents.
+- The scraper respects rate limits (2s delay between requests) — do not reduce this.
+- Python version: 3.12. Dependencies are pinned in `requirements.txt`.