// Run a local pnpm monorepo CI loop, fix failures, and stop only when the full sequence is green. Use when Codex is asked to fix CI, make a repository green, or resolve failures from `pnpm -w build`, `pnpm -w test:type`, `pnpm -w lint`, `pnpm -w test`, or `pnpm -w test:integration`.
Use when working with Hot Updater CLI setup, deployment, patch artifacts, bundle inventory/state, rollback, channels, code signing keys, database migration, diagnostics, or AI-assisted React Native OTA operations.
Generate and execute PR-aware OTA E2E scenarios for `examples/v0.85.0` by diffing the checked-out branch against its PR base branch or default branch, inferring the affected runtime, rollout, and recovery surfaces, and then running the scenario through `../e2e`. Use when the caller wants current-branch OTA validation without hand-writing the scenario.
Run a fixed OTA regression flow for `examples/v0.85.0` with `agent-device`. Use when the caller wants a built-in end-to-end scenario instead of defining one manually: deploy a known-good OTA bundle, verify a visible UI change plus the deployed `bundleId`, then deploy an intentionally crashing OTA bundle and verify rollback to the previous stable bundle with `RECOVERED` and `crashedBundleId` evidence.
Run end-to-end OTA verification for `examples/v0.85.0` with `agent-device`. Use when validating iOS or Android release builds, deploying OTA bundles with `pnpm hot-updater deploy`, checking stable update application, reproducing rollback after a crash bundle, or reading bundle-store metadata and crash history for the v0.85.0 example app.
Use when an AI agent needs to queue, monitor, diagnose, and iterate on HotUpdater E2E dashboard jobs through the Mac mini `hot-updater-agent` CLI instead of running local device flows directly. Trigger for requests to run PR E2E through the dashboard, inspect queued E2E failures, tail bot logs, or fix-and-rerun E2E until green.
Automates interactions for iOS simulators/devices and Android emulators/devices. Use when navigating apps, taking snapshots/screenshots, tapping, typing, scrolling, or extracting UI info on mobile targets.
| name | fix-ci |
| description | Run a local pnpm monorepo CI loop, fix failures, and stop only when the full sequence is green. Use when Codex is asked to fix CI, make a repository green, or resolve failures from `pnpm -w build`, `pnpm -w test:type`, `pnpm -w lint`, `pnpm -w test`, or `pnpm -w test:integration`. |
Use this skill to drive a repository from red to green with the exact
command order pnpm -w build, pnpm -w test:type, pnpm -w lint,
pnpm -w test, pnpm -w test:integration. Fix root causes in code,
config, or tests and keep iterating until the full ordered run passes.
git status --short so you do not
overwrite unrelated user changes..agents/skills/fix-ci/scripts/run_fixci.sh all. The script stops at
the first failing step and writes logs under .codex/fix-ci/<timestamp>/
by default.|| true, blanket rule disables, test.skip, or broad
type suppressions unless the user explicitly asks for that tradeoff..agents/skills/fix-ci/scripts/run_fixci.sh <step> until it passes.
Supported single steps are build, test:type, type, typecheck,
lint, test, test:integration, and integration.build; otherwise continue from the step
that failed..agents/skills/fix-ci/scripts/run_fixci.sh all
succeeds end-to-end..agents/skills/fix-ci/scripts/run_fixci.sh: Run one step or the full
suite with timestamped logs.