Thanks for helping improve Memory MCP for long-term AI project memory and Supabase persistent context.
- Fork the repository and create a feature branch.
- Install development dependencies with
pip install -r requirements-dev.txt. - Run
pytest,black .,flake8, andmypy srcbefore opening a pull request. - Document any user-facing change in
README.mdordocs/.
- Use Python type hints in every function.
- Keep lines at 100 characters or fewer.
- Prefer small functions and explicit error handling.
- Write self-documenting code and only add comments when logic is not obvious.
- Keep docs bilingual whenever a public page or message changes.
- Explain the problem, the solution, and any schema or API impact.
- Include screenshots or documentation updates when the docs UI changes.
- Add or update tests for every server tool, optimizer behavior, or SQL contract you modify.
- Wait for review and keep commits focused so maintainers can audit changes quickly.
Open issues with this structure:
Title: Short, searchable summary
Problem
- What failed?
- Which interface was involved?
Expected behavior
- What should happen?
Environment
- Python version
- Interface: native / qwen-code / codex
- Supabase setup notes
Extra context
- Logs, screenshots, sample payloads, or schema details