Runtime enforcement boundary for AI agents: a local sidecar that gates every outbound call against Cedar policies you own. Deterministic, call-level, no model on the hot path
These files are Firma-AI/openfirma's own configuration. They tell Claude Code, Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.
AGENTS.md A 2,310 tok CLAUDE.md A 3 tok .skills/adversarial-review/SKILL.md A 44 tok .skills/commit-guidelines/SKILL.md A — .skills/docs-guidelines/SKILL.md A 19 tok .skills/edit-pr/SKILL.md A 38 tok .skills/fix-jj-conflicts/SKILL.md A 33 tok .skills/lint/SKILL.md A 19 tok .skills/open-pr/SKILL.md A 23 tok .skills/planning-changes/SKILL.md A 47 tok .skills/review-rust-code/SKILL.md A 31 tok .skills/reviewing-changes/SKILL.md A 34 tok .skills/reviewing-plans/SKILL.md A 44 tok .skills/run-rust-tests/SKILL.md A 20 tok .skills/rust-docs-guidelines/SKILL.md A 16 tok .skills/rust-tests-guidelines/SKILL.md A 15 tok .skills/split-jj-changeset/SKILL.md A — .skills/verify/SKILL.md A 27 tok .skills/write-rust-tests/SKILL.md A 17 tok .skills/writing-black-box-tests/SKILL.md A 39 tok