Versioned Constitution Governance
Problem
When an agent rewrites its own "constitution," it may accidentally violate safety or regress on alignment objectives if changes aren't reviewed.
Solution
Store the constitution in a version-controlled, signed repository:
- YAML/TOML rules live in Git.
- Each commit is signed (e.g., Sigstore); CI runs automated policy checks.
- Only commits signed by approved reviewers or automated tests are merged.
- The agent can propose changes, but a gatekeeper merges them.
How to use it
- Require
git commit -Sor similar. - Run diff-based linting to flag deletions of critical rules.
- Expose constitution
HEADas read-only context in every agent episode.
References
- Hiveism, Self-Alignment by Constitutional AI
- Anthropic, Constitutional AI white-paper