fix(test): skip live auth browser caches#91907
Conversation
|
Codex review: needs maintainer review before merge. Reviewed June 30, 2026, 5:03 PM ET / 21:03 UTC. Summary PR surface: Tests +62. Total +62 across 2 files. Reproducibility: yes. source-level. With Review metrics: 1 noteworthy metric.
Stored data model Root-cause cluster Members:
Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything. Merge readiness Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch. Rank-up moves:
Risk before merge
Maintainer options:
Next step before merge
Security Review detailsBest possible solution: Land this scoped live-test staging filter after maintainer acceptance of the all-root scope, or narrow it to Do we have a high-confidence way to reproduce the issue? Yes, source-level. With Is this the best way to solve the issue? Yes, with one maintainer scope choice. The patch fixes the implicated live-test staging layer, preserves known auth/config files, leaves core runtime state copying alone, and can be narrowed to AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against b885c81479d0. Label changesLabel justifications:
Evidence reviewedPR surface: Tests +62. Total +62 across 2 files. View PR surface stats
What I checked:
Likely related people:
What the crustacean ranks mean
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics. How this review workflow works
Review history (1 earlier review cycle)
|
b486b4c to
65e7a5e
Compare
|
Maintainer verification for What changed during review:
Dependency contract checked directly against Gemini CLI Proof:
No remaining proof gaps. Land-ready. |
|
Merged via squash.
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 65e7a5e159
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| const filter = | ||
| authDir === ".gemini" | ||
| ? (entryPath: string) => shouldStageLiveGeminiPath(sourcePath, entryPath) | ||
| : undefined; | ||
| copyDirIfExists(sourcePath, path.join(params.tempHome, authDir), { filter }); |
There was a problem hiding this comment.
In Docker live contexts this filter never runs: google-gemini-cli still resolves to the whole .gemini auth dir in scripts/lib/live-docker-auth.sh:105-107, and the Docker harness copies that mounted dir with cp -R (for example scripts/test-live-gateway-models-docker.sh:142-145; cp --help documents -R as recursively copying directories). As a result, pnpm test:docker:live-gateway:gemini/default Docker live runs can still copy the same Antigravity browser profiles into the container or prestaged home and fill disk before Vitest reaches this local installTestEnv staging path.
Useful? React with 👍 / 👎.
* fix(test): skip live auth browser caches * fix(test): scope Gemini live auth exclusions --------- Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(test): skip live auth browser caches * fix(test): scope Gemini live auth exclusions --------- Co-authored-by: Peter Steinberger <steipete@gmail.com>
Summary
.gemini..gemini/oauth_creds.jsonand.gemini/settings.json..openclaw/credentialsand plugin state copying are unchanged.Linked context
Closes #91893
Related #91893
Was this requested by a maintainer or owner? No.
Real behavior proof (required for external PRs)
.geminibrowser cache/profile/recording trees while preserving Gemini auth files.OPENCLAW_LIVE_TEST=1, using a temporary HOME containing.geminiauth files plus cache-like browser directories.node --import tsx --input-type=modulewith a temp HOME fixture, importinginstallTestEnvfromtest/test-env.ts, then checking the staged temp HOME.{ "tempHomeDiffers": true, "copiedGeminiOauth": true, "copiedGeminiSettings": true, "skippedBrowserProfile": true, "skippedRecordings": true, "skippedGpuCache": true, "skippedServiceWorkerCache": true }origin/mainhascopyDirIfExistsusing recursivefs.cpSyncwith no filter for.geministaging.Tests and validation
node scripts/run-vitest.mjs test/test-env.test.tspassed, 4 tests.pnpm exec oxfmt --check --threads=1 test/test-env.ts test/test-env.test.tspassed.pnpm changed:lanes --jsonselected onlytooling.git diff --checkpassed.OPENCLAW_CHECK_CHANGED_REMOTE_CHILD=1 OPENCLAW_CHANGED_LANES_RAW_SYNC=1 CI=1 PNPM_CONFIG_VERIFY_DEPS_BEFORE_RUN=false corepack pnpm check:changedpassed.Direct
pnpm check:changedcould not start because the local Crabbox binary failed its basic sanity check before running the gate.Regression coverage adds
.geminiauth files plus generated browser cache/profile fixtures to the existing live HOME staging test.Risk checklist
Did user-visible behavior change? No.
Did config, environment, or migration behavior change? No.
Did security, auth, secrets, network, or tool execution behavior change? No.
What is the highest-risk area? Over-filtering live test auth staging.
How is that risk mitigated? The filter applies only to external auth directory staging and the regression test proves Gemini auth/settings files still copy.
Current review state
What is the next action? Maintainer review.
What is still waiting on author, maintainer, CI, or external proof? CI and ClawSweeper review.
Which bot or reviewer comments were addressed? None yet.