project-config-mutate

To atomically mutate PROJECT_CONFIG.md, apply targeted add or remove operations through a review gate that enforces target-type validation, generates diffs, and uses a temporary file protocol to ensure safe writes and post-write cache regeneration.

Safe ai-workflow config changes require an atomic write protocol that validates domain scoping and target-types, generates a diff for traceability, and regenerates caches post-write to prevent state corruption.

The project-config-review diff generation process works by applying a single add or remove operation to PROJECT_CONFIG.md, validating the value shape and domain scoping, then presenting the resulting diff for review before committing the atomic write.

Yes, you can remove domain-scoped target-types from PROJECT_CONFIG without idempotency issues because the mutation process enforces idempotency checks, ensuring repeated remove operations do not corrupt the config or trigger duplicate errors.

Modifying project config requires regenerating the cache after writing to ensure the ai-workflow environment immediately reflects the atomic write and domain-scoped mutations without stale data.

Project config mutation fails validation before the review gate when the target-type or value shape does not match domain declarations, preventing invalid configurations from reaching the atomic write stage.