The 3.10.0 milestone was about making the advertised smoke pack trustworthy enough to act like a real release gate. The main drift was in the repro-plus-fix scenario: the recipe docs were SDK-first, but the smoke still shelled out to CLI patch apply and asserted a human summary string.\n\nSwitch the smoke runner to use the structured SDK patch flow directly, remove the harness-only CLI dependency, and tighten the fake smoke tests so they prove the same structured path the docs recommend. This keeps smoke failures tied to real user-facing regressions instead of human-output formatting drift.\n\nPromote make smoke-use-cases as the trustworthy guest-backed verification path in the top-level docs, bump the release surface to 3.10.0, and mark the roadmap milestone done.\n\nValidation:\n- uv lock\n- UV_CACHE_DIR=.uv-cache uv run pytest --no-cov tests/test_workspace_use_case_smokes.py\n- UV_CACHE_DIR=.uv-cache make check\n- UV_CACHE_DIR=.uv-cache make dist-check\n- USE_CASE_ENVIRONMENT=debian:12 UV_CACHE_DIR=.uv-cache make smoke-use-cases
1.6 KiB
1.6 KiB
Workspace Use-Case Recipes
These recipes turn the stable workspace surface into five concrete agent flows. They are the canonical next step after the quickstart in install.md or first-run.md.
Run all real guest-backed scenarios locally with:
make smoke-use-cases
Recipe matrix:
| Use case | Recommended profile | Smoke target | Recipe |
|---|---|---|---|
| Cold-start repo validation | workspace-full |
make smoke-cold-start-validation |
cold-start-repo-validation.md |
| Repro plus fix loop | workspace-core |
make smoke-repro-fix-loop |
repro-fix-loop.md |
| Parallel isolated workspaces | workspace-core |
make smoke-parallel-workspaces |
parallel-workspaces.md |
| Unsafe or untrusted code inspection | workspace-core |
make smoke-untrusted-inspection |
untrusted-inspection.md |
| Review and evaluation workflows | workspace-full |
make smoke-review-eval |
review-eval-workflows.md |
All five recipes use the same real Firecracker-backed smoke runner:
uv run python scripts/workspace_use_case_smoke.py --scenario all --environment debian:12
That runner generates its own host fixtures, creates real guest-backed workspaces,
verifies the intended flow, exports one concrete result when relevant, and cleans
up on both success and failure. Treat make smoke-use-cases as the trustworthy
guest-backed verification path for the advertised workspace workflows.