Skip to content

Contributing

People and AI agents work on ritueel by the same rules. The rules are written once, on the pages below; the files in the repository point here instead of repeating them.

Who decides what is in GOVERNANCE.md. How to propose a feature or a decision is in CONTRIBUTING.md. A suspected vulnerability goes through SECURITY.md, never through an issue or a pull request. Questions and ideas are welcome as issues.

Read the page for what you are about to do, then do it.

Before you Read
create a branch, commit, push, open a pull request or answer a review Branches and pull requests
run a build, a test or a check, or add a tool Tools, checks and verification
propose, write or rewrite a decision record Decision records
run a spike or write up research Spikes and research
create or edit a page of this site Pages and Writing style
write a name, an address, an id or anything secret Public content and secrets
contribute with an AI agent, or run a team of agents AI agents
  • CONTRIBUTING.md: the entry for a new contributor, and how to propose a feature or a decision.
  • GOVERNANCE.md: roles, how a decision is made, review rules and the release numbering.
  • SECURITY.md: how to report a vulnerability privately.
  • CHANGELOG.md: what changed for the person who uses a release.
  • AGENTS.md: the short form of these pages for an AI agent, with the page to read before each kind of change. CLAUDE.md loads it for Claude Code.
  • .claude/skills/ and .claude/agents/: task and role wrappers for Claude Code that point back to these pages. They hold no rules of their own.
  • docs/decisions/README.md: the format of the decision records.

When a repository file and a page disagree, the page is the rule and the file is fixed.