The workflow directory mixed entry points, reusable stages and scheduled jobs into one flat list sorted by topic, so the first question about any file - what starts it - was the one the name did not answer. Every workflow now carries one of three prefixes, derived from its on: block: cron- for a schedule trigger, call- for one only reachable through workflow_call, entry- for everything a repository event or a human starts. Scheduled workflows additionally carry an alarm clock as the first character of their name. Two lints keep it that way. One derives the expected prefix from the triggers and compares it against the file name; it flagged all 17 offenders before the rename and none after. The other resolves every .github/workflows/<file> string in the repository against the directory, which is what a rename silently breaks in Markdown links, Python constants and shell API URLs alike. Renaming pulled 69 pre-existing comments into the comment lint's changed-file scope. The vendored CodeQL template got the nocheck marker its scorecard sibling already carried, two workflow headers moved above name: where they count as file headers, and the remaining banners were deleted. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
12 KiB
CI Configuration ⚙️
This page lists every repository variable that controls GitHub Actions CI behaviour in this repo. For the workflow catalog see workflows.md; for the CI flow see pipeline.md.
Repository variables are set under Settings → Secrets and variables → Actions → Variables.
Variables 📋
| Variable | Workflow | Default (unset) | Set to activate |
|---|---|---|---|
CI_CANCEL_IN_PROGRESS |
entry-push-latest.yml, call-orchestrator.yml | Cancels in-progress runs on new push | false to keep in-progress runs alive |
CI_SYNC_MAIN_SOURCE_REPOSITORY |
entry-push-latest.yml | Syncs main from infinito-nexus/core before CI scope discovery |
<owner>/<repo> to use another source, or false, empty, or the current repository to skip |
CI_RUN_ON_MAIN |
entry-push-latest.yml | Pushes to main skip CI |
true to run CI on main pushes too |
CI_ENABLE_AUTO_UPDATES |
cron-update.yml, entry-pr-open-dependabot-close.yml | Update jobs skipped; Dependabot PRs auto-closed | true to allow update PRs (workflow-driven and Dependabot) |
INFINITO_PLAYWRIGHT_KEEP |
call-test-deploy-compose.yml | Playwright keeps trace, screenshot and video only when a test fails | true to keep them for every test (passing runs included) |
CI_RERUN_INTERVAL_MINUTES |
rerun-cancelled.yml | Sweeps for cancelled runs every full hour | Minutes between two sweeps, rounded up to whole hours |
CI_RERUN_MAX_ATTEMPTS |
rerun-cancelled.yml | Revives a run however often it is cancelled | Attempt count at which a run is left alone; 0 for no ceiling |
CI_RERUN_MAX_AGE_HOURS |
rerun-cancelled.yml | Ignores runs untouched for more than 24 hours | Age in hours beyond which a cancelled run is left alone |
CI_CANCEL_IN_PROGRESS 🛑
Controls whether a new run cancels an already-running CI pipeline on the same ref. Applies to the push entry workflow and to the orchestrator's own concurrency group, which every entry workflow shares. The manual entry workflow overrides the variable on both levels: its own group is cancel-in-progress: true, and it passes force_cancel_in_progress: true into the orchestrator.
Default behaviour (variable not set or set to any value other than false):
In-progress runs are cancelled when a new push arrives. This is the recommended setting for most workflows.
To disable cancellation:
- Open the repository on GitHub.
- Go to Settings → Secrets and variables → Actions.
- Switch to the Variables tab.
- Click New repository variable.
- Set Name to
CI_CANCEL_IN_PROGRESSand Value tofalse. - Save.
To re-enable cancellation:
Delete the variable or change its value to anything other than false (e.g. true).
How it works:
cancel-in-progress: ${{ vars.CI_CANCEL_IN_PROGRESS != 'false' }}
| Variable value | Expression result | Behaviour |
|---|---|---|
| (not set / empty) | '' != 'false' → true |
Cancels in-progress runs ✓ |
false |
'false' != 'false' → false |
Does not cancel ✓ |
true |
'true' != 'false' → true |
Cancels in-progress runs ✓ |
CI_SYNC_MAIN_SOURCE_REPOSITORY 🔄
Controls whether the push workflow updates the current repository's main branch before any later job derives deploy scope from origin/main.
This keeps fork branch pushes aligned with the canonical default branch so diff-driven role discovery sees only the branch's own changes.
⚠️ Forced overwrite — do not work in
main. When the sync runs, the current repository'smainis force-pushed from the source repository'smain. Any commit that exists only on the fork'smainis discarded without warning on the next push to any covered branch. Never commit or merge work directly intomainon a synced repository. Always work on afeature/**,fix/**, orhotfix/**branch; treatmainas a read-only mirror of the source. If you must usemainas a working branch, disable the sync first (see To disable the sync below).
Default behaviour (variable not set):
The workflow uses infinito-nexus/core as the source repository.
When the current repository is already infinito-nexus/core, the same-repository skip applies.
To use a different source repository:
- Open the repository on GitHub.
- Go to Settings → Secrets and variables → Actions.
- Switch to the Variables tab.
- Click New repository variable.
- Set Name to
CI_SYNC_MAIN_SOURCE_REPOSITORYand Value to<owner>/<repo>. - Save.
To disable the sync:
Set the variable to false, 0, no, off, none, an empty value, or the current repository name.
| Variable value | Behaviour |
|---|---|
| (not set) | Uses infinito-nexus/core; syncs forks and skips in infinito-nexus/core itself ✓ |
infinito-nexus/core in a fork |
Syncs the fork's main from infinito-nexus/core ✓ |
| current repository | Sync skipped ✓ |
false, 0, no, off, none, or empty |
Sync skipped ✓ |
CI_RUN_ON_MAIN 🎯
Controls whether pushes to main trigger the CI pipeline. Pushes to all other branches covered by the workflow (feature/**, hotfix/**, fix/**, alert-autofix-*) are unaffected.
Default behaviour (variable not set or set to any value other than true):
Pushes to main are gated out at the run-policy job and CI is skipped.
To enable CI on main pushes:
- Open the repository on GitHub.
- Go to Settings → Secrets and variables → Actions.
- Switch to the Variables tab.
- Click New repository variable.
- Set Name to
CI_RUN_ON_MAINand Value totrue. - Save.
To disable again:
Delete the variable or change its value to anything other than true.
How it works:
The gate is applied inside push_ci_policy.sh, which the run-policy job invokes. When GITHUB_REF == refs/heads/main and CI_RUN_ON_MAIN != 'true', the job emits should_run=false and every downstream job is skipped.
| Variable value | Ref is main |
Behaviour |
|---|---|---|
| (not set / empty) | yes | CI skipped ✓ |
true |
yes | CI runs ✓ |
| any other value | yes | CI skipped ✓ |
| (any) | no | Unaffected (CI runs per branch rules) ✓ |
CI_ENABLE_AUTO_UPDATES 🔄
Controls whether automated update PRs are created. Covers both the workflow-driven jobs in cron-update.yml (Docker image versions, agent skills) and PRs opened by Dependabot (gated via entry-pr-open-dependabot-close.yml, which auto-closes them).
The workflow-driven jobs additionally require the BOT_APP_CLIENT_ID and BOT_APP_PRIVATE_KEY repository secrets, see secrets.md. Without those secrets, runs fail at the token-minting step and no PR is opened.
Default behaviour (variable not set or set to any value other than true):
The update-docker-image-versions job is skipped. Dependabot PRs are auto-closed on open with a comment pointing to this variable.
To enable update PRs:
- Open the repository on GitHub.
- Go to Settings → Secrets and variables → Actions.
- Switch to the Variables tab.
- Click New repository variable.
- Set Name to
CI_ENABLE_AUTO_UPDATESand Value totrue. - Save.
To disable again:
Delete the variable or change its value to anything other than true.
How it works:
In cron-update.yml each job carries a job-level guard:
if: vars.CI_ENABLE_AUTO_UPDATES == 'true'
Dependabot cannot read repository variables itself, so entry-pr-open-dependabot-close.yml listens on pull_request_target (opened, reopened) and closes any PR authored by dependabot[bot] while CI_ENABLE_AUTO_UPDATES != 'true'. The workflow does not check out PR code, which keeps the elevated pull_request_target context safe.
| Variable value | Workflow update jobs | Dependabot PRs |
|---|---|---|
| (not set / empty) | Skipped ✓ | Auto-closed on open ✓ |
true |
Run ✓ | Stay open ✓ |
| any other value | Skipped ✓ | Auto-closed on open ✓ |
INFINITO_PLAYWRIGHT_KEEP 🎬
Controls whether Playwright keeps trace, screenshot, and video for every test or only for failing tests in the compose deploy-test workflow (call-test-deploy-compose.yml). For the full propagation chain, the inventory override, and the local equivalents, see Playwright Tests.
Default behaviour (variable not set or set to any value other than true):
Artefacts are retained only when a test fails.
To retain artefacts for every test:
- Open the repository on GitHub.
- Go to Settings → Secrets and variables → Actions.
- Switch to the Variables tab.
- Click New repository variable.
- Set Name to
INFINITO_PLAYWRIGHT_KEEPand Value totrue. - Save.
To disable again:
Delete the variable or change its value to anything other than true.
How it works:
Each deploy-test workflow forwards the variable into its own env: block:
INFINITO_PLAYWRIGHT_KEEP: ${{ vars.INFINITO_PLAYWRIGHT_KEEP }}
| Variable value | Behaviour |
|---|---|
| (not set / empty) | Trace / screenshot / video kept only on failure ✓ |
true |
Trace / screenshot / video kept for every test ✓ |
| any other value | Trace / screenshot / video kept only on failure ✓ |
CI_RERUN_* 🔁
Three variables tune rerun-cancelled.yml, the sweep that restarts the failed jobs of a branch whose latest run was cancelled.
| Variable | Unset | Effect when set |
|---|---|---|
CI_RERUN_INTERVAL_MINUTES |
60 |
Minutes between two sweeps |
CI_RERUN_MAX_ATTEMPTS |
0 |
A run at or above this attempt count is left alone; 0 means no ceiling, a run is revived however often it is cancelled |
CI_RERUN_MAX_AGE_HOURS |
24 |
A run untouched for longer than this is left alone, so a sweep does not resurrect abandoned branches |
How the interval works:
A cron expression cannot read a repository variable, so the workflow is triggered every full hour and a gate lets only the hours that open the configured window through:
INTERVAL_MINUTES: ${{ vars.CI_RERUN_INTERVAL_MINUTES || '60' }}
The interval is rounded up to whole hours and counted from 00:00 UTC, so 60 sweeps every hour, 180 at 00:00, 03:00, 06:00 and so on, and 1440 once a day at 00:00. One hour is the floor: anything below it sweeps hourly. Anything above 24 hours collapses to a daily sweep. Manual dispatch skips the gate entirely and always sweeps.
The gate reads the hour, not the minute, so a scheduled run that GitHub starts a few minutes late still counts for its hour.
On the missing ceiling: with CI_RERUN_MAX_ATTEMPTS unset, a run that is cancelled again after every revival is revived again, indefinitely. CI_RERUN_MAX_AGE_HOURS does not bound this, because each new attempt refreshes the run's updatedAt and so keeps it inside the age window. Set the variable to a positive number if a run should eventually be given up on.