Skip to content

CLAUDE.md

Latest commit

 

History

History
24 lines (18 loc) · 1.59 KB

CLAUDE.md

File metadata and controls

24 lines (18 loc) · 1.59 KB

Claude Code Instructions for SecBrain

After completing any task

  1. Always commit your work. Stage all relevant changes and create a commit with a clear message following the project's <type>: <description> format (feat, fix, docs, style, refactor, test, chore).
  2. Always push after committing. Keep the remote in sync — run git push after every commit. The repo should never be ahead of origin when a task is done.
  3. Clean up temporary files. Remove any scratch files, temp outputs, or artifacts created during the session (e.g., commitmsg.txt, leftover debug files).
  4. Clean up branches. If you created a feature branch and the work is merged or complete, delete it locally and remotely. The default workflow is to work directly on main unless the user says otherwise.

Platform and environment

  • This project is developed on Windows but tools like Foundry, nuclei, and shell scripts target Linux/WSL.
  • Always use pathlib.Path for file operations, never os.path.
  • Use Path.expanduser() when resolving paths that contain ~.
  • Use shutil.which() to verify tool availability before calling external executables.
  • Never hardcode platform-specific paths (/usr/bin/, C:\, /mnt/c/).
  • The repo has a .gitattributes enforcing LF line endings for scripts and source files. Do not override this.

Code standards

  • Python 3.11+, strict mypy typing, ruff linting (line length 100).
  • Use structlog for logging, not print statements.
  • Follow existing patterns in the codebase for new code.
  • Tests go in secbrain/tests/. Run with: cd secbrain && pytest tests/ -v