test_review_live.py (M7.3) has never run in CI. scripts/e2e.sh ran only tests/integration/test_end_to_end_loop.py, and the integration step runs -m "integration and not e2e", which skips it. Found while checking that the NISSE_REVIEW_GITEA_TOKEN secret works: manual run 614 passed with 9 e2e tests, none of them the review test.
Changes
scripts/e2e.sh runs every e2e-marked test under tests/integration/. Today that's the loop's tests plus the review test.
test_review_live.py is now smoke too, so every PR runs it. It takes about a minute on the local model, and it's the only check that the review account, its token and the read-only refusal still work.
The Development Plan's M7.3 notes say so.
Verification
pytest -m "e2e and smoke" --collect-only tests/integration collects 4 tests: the 3 loop smoke tests plus the review test.
This PR is labelled e2e:full, so its CI run uses the new secret for the first time.
`test_review_live.py` (M7.3) has never run in CI. `scripts/e2e.sh` ran only `tests/integration/test_end_to_end_loop.py`, and the integration step runs `-m "integration and not e2e"`, which skips it. Found while checking that the `NISSE_REVIEW_GITEA_TOKEN` secret works: manual run 614 passed with 9 e2e tests, none of them the review test.
**Changes**
- `scripts/e2e.sh` runs every `e2e`-marked test under `tests/integration/`. Today that's the loop's tests plus the review test.
- `test_review_live.py` is now `smoke` too, so every PR runs it. It takes about a minute on the local model, and it's the only check that the review account, its token and the read-only refusal still work.
- The Development Plan's M7.3 notes say so.
**Verification**
- `pytest -m "e2e and smoke" --collect-only tests/integration` collects 4 tests: the 3 loop smoke tests plus the review test.
- This PR is labelled `e2e:full`, so its CI run uses the new secret for the first time.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
https://claude.ai/code/session_01J5Utyfzw7gF7UhjRDEtjTk
scripts/e2e.sh ran only test_end_to_end_loop.py, and the integration step
excludes e2e-marked tests, so test_review_live.py (M7.3) never ran in CI.
The script now runs every e2e-marked test under tests/integration, and the
review test is a smoke test: about a minute on the local model, and the only
check that the review account, its token and its read-only refusal work.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01J5Utyfzw7gF7UhjRDEtjTk
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
test_review_live.py(M7.3) has never run in CI.scripts/e2e.shran onlytests/integration/test_end_to_end_loop.py, and the integration step runs-m "integration and not e2e", which skips it. Found while checking that theNISSE_REVIEW_GITEA_TOKENsecret works: manual run 614 passed with 9 e2e tests, none of them the review test.Changes
scripts/e2e.shruns everye2e-marked test undertests/integration/. Today that's the loop's tests plus the review test.test_review_live.pyis nowsmoketoo, so every PR runs it. It takes about a minute on the local model, and it's the only check that the review account, its token and the read-only refusal still work.Verification
pytest -m "e2e and smoke" --collect-only tests/integrationcollects 4 tests: the 3 loop smoke tests plus the review test.e2e:full, so its CI run uses the new secret for the first time.🤖 Generated with Claude Code
https://claude.ai/code/session_01J5Utyfzw7gF7UhjRDEtjTk