The starter adds policy and evidence files. It does not replace the engine, build system, Claude configuration, Codex configuration, or repository tests.
repository-template/ to the game repository.profiles/game.json to .ai-safe2/profiles/game.json..ai-safe2/workflow.json.Validate the package and generate the repository-specific explanation:
python scripts/validate_workflow.py --root .
python scripts/explain_workflow.py \
--workflow .ai-safe2/workflow.json \
--profile .ai-safe2/profiles/game.json \
--output AI-SAFE2-DEVELOPER-WORKFLOW.md
AI-SAFE2-DEVELOPER-WORKFLOW.md and confirm the routes, trust
boundaries, owner, review options, and limitations match the game.The starter GitHub workflow verifies the pinned CLI, creates the deterministic
route, runs CLI decision routing, initializes the revision-bound human decision
record to hold, repeats the repository explanation, and uploads those
artifacts. A green validation proves that these required records were produced
and that the configuration is internally consistent. It does not prove that an
optional AI reviewer ran or that the game is secure or ready to release.
| Before | After |
|---|---|
| The same review habit may be used for every change | Documentation, gameplay code, and critical trust-boundary changes can take different routes |
| Claude, Codex, and scanners can produce disconnected comments | Results can be retained as attributed, revision-bound evidence |
| A missing reviewer may be easy to overlook | Failed, unavailable, and stale evidence do not satisfy a required capability |
| “Checks passed” can sound like release approval | The workflow stops at ready_for_human_decision; a named human decides |
Start with the route and deterministic checks. Add planning, semantic review, security review, or runtime governance only where the repository’s risks and goals justify them. See the Adoption Guide for the module menu and tradeoffs.
Use a small, reversible change with tests. The pull request should demonstrate:
Do not start with payment, authentication, multiplayer authority, build signing, or destructive save migration.