8 of 41 repositories we own do not have a genuinely green default branch.
cancelled, timed_out and startup_failure are not green: such a run proves nothing about the tree. The usual cause is concurrency cancel-in-progress when several merges land at once, so re-run the cancelled run rather than assuming the older success still stands.
This issue is rewritten in place once a day and closed when everything is green. Editing a body does not notify, so a long-lived gap does not become a daily ping.
- Platform manifest and source boundary — newest run concluded failure on
927a721 (run)
- Release and PyPI Publish: stale-green — success, but it tested
2aa387f and head is 927a721 -- that pass is evidence about a tree nobody runs any more
- Draft post (scan, classify, author) — newest run concluded failure on
bc8623b (run)
- Release and PyPI Publish — newest run concluded failure on
2a3ef96; the workflow file changed afterwards (2026-09-03T20:16:39Z), so re-run it to find out whether the fix took (run)
- FFmpeg pin: stale-green — success, but it tested
fb46907 and head is c458549 -- that pass is evidence about a tree nobody runs any more
- Native Installer Release — newest run concluded cancelled on
7c59750 AFTER the workflow was retired (2026-08-29T14:52:58Z), so it failed as currently configured (run)
- Production channel admission — newest run concluded failure on
9a1708e (run)
- Build managed FFmpeg runtime: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-19T20:13:07Z, after that run)
- Notify docs: stale-green — success, but it tested
095082c and head is 9a1708e -- that pass is evidence about a tree nobody runs any more
- evidence-freshness — newest run concluded failure on
38ebe51 (run)
- Source policy sync — newest run concluded failure on
0ec1734 (run)
- Qualification evidence store: stale-green — success, but it tested
e75cf05 and head is 0ec1734 -- that pass is evidence about a tree nobody runs any more
- Production DB backup freshness — newest run concluded failure on
660f040 (run)
- Production DB logical backup — newest run concluded failure on
660f040 (run)
- Published version claims — newest run concluded failure on
660f040 (run)
- Action pin sweep: stale-green — success, but it tested
c827d80 and head is 660f040 -- that pass is evidence about a tree nobody runs any more
- Default branch sweep: in-flight — in_progress for 0.0h on
660f040 -- not yet evidence either way
- Docs CI: stale-green — success, but it tested
7045af6 and head is 660f040 -- that pass is evidence about a tree nobody runs any more
- CodeQL — newest run concluded failure on
53eabff (run)
- Published claims — newest run concluded failure on
53eabff (run)
Green, with notes — none of these is a failure
retired and not-push-gated entries are frozen history: GitHub keeps a workflow's last run for ever, so a workflow that stopped running would otherwise be reported red permanently. stale-green passed on an older tree. in-flight has been running for less than 6h.
- OpenAdaptAI/openadapt-cloud
- Dependency Audit: stale-green — success, but it tested
60df757 and head is 8a0ec08 -- that pass is evidence about a tree nobody runs any more
- Execute live synthetic acceptance: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-19T22:44:59Z, after that run)
- Requalify managed synthetic workflow: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-29T19:45:20Z, after that run)
- Secret Scan: stale-green — success, but it tested
74c6e65 and head is 8a0ec08 -- that pass is evidence about a tree nobody runs any more
- OpenAdaptAI/openadapt-flow
- Paper: stale-green — success, but it tested
dad7b04 and head is 3b40a9a -- that pass is evidence about a tree nobody runs any more
- Release and PyPI Publish: stale-green — success, but it tested
6d1f80a and head is 3b40a9a -- that pass is evidence about a tree nobody runs any more
- Validate claims: stale-green — success, but it tested
98588e8 and head is 3b40a9a -- that pass is evidence about a tree nobody runs any more
- OpenAdaptAI/openadapt-grounding
- Release and PyPI Publish: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-20T11:10:24Z, after that run)
- OpenAdaptAI/openadapt-herald
- weekly-digest: retired — workflow disabled_inactivity; the failure run is frozen history
- OpenAdaptAI/openadapt-retrieval
- Release and PyPI Publish: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-20T11:10:33Z, after that run)
- OpenAdaptAI/openadapt-viewer
- Generate Screenshots: stale-green — success, but it tested
01e87e5 and head is 6cfad96 -- that pass is evidence about a tree nobody runs any more
- Release and PyPI Publish: not-push-gated — no change or schedule trigger, so the failure run says nothing about the tree (retired 2026-08-20T11:10:42Z, after that run)
- OpenAdaptAI/openadapt-wright
- Deploy Worker: stale-green — success, but it tested
54311b1 and head is 21ec989 -- that pass is evidence about a tree nobody runs any more
Swept 41 repositories in 217 API reads. https://github.com/OpenAdaptAI/openadapt-ops/actions/runs/33962789306
8 of 41 repositories we own do not have a genuinely green default branch.
cancelled,timed_outandstartup_failureare not green: such a run proves nothing about the tree. The usual cause is concurrencycancel-in-progresswhen several merges land at once, so re-run the cancelled run rather than assuming the older success still stands.This issue is rewritten in place once a day and closed when everything is green. Editing a body does not notify, so a long-lived gap does not become a daily ping.
OpenAdaptAI/OpenAdapt (
main)927a721(run)2aa387fand head is927a721-- that pass is evidence about a tree nobody runs any moreOpenAdaptAI/openadapt-blog (
main)bc8623b(run)OpenAdaptAI/openadapt-capture (
main)2a3ef96; the workflow file changed afterwards (2026-09-03T20:16:39Z), so re-run it to find out whether the fix took (run)fb46907and head isc458549-- that pass is evidence about a tree nobody runs any moreOpenAdaptAI/openadapt-desktop (
main)7c59750AFTER the workflow was retired (2026-08-29T14:52:58Z), so it failed as currently configured (run)9a1708e(run)095082cand head is9a1708e-- that pass is evidence about a tree nobody runs any moreOpenAdaptAI/openadapt-evals (
main)38ebe51(run)OpenAdaptAI/openadapt-internal (
main)0ec1734(run)e75cf05and head is0ec1734-- that pass is evidence about a tree nobody runs any moreOpenAdaptAI/openadapt-ops (
main)660f040(run)660f040(run)660f040(run)c827d80and head is660f040-- that pass is evidence about a tree nobody runs any more660f040-- not yet evidence either way7045af6and head is660f040-- that pass is evidence about a tree nobody runs any moreOpenAdaptAI/openadapt-web (
main)53eabff(run)53eabff(run)Green, with notes — none of these is a failure
retiredandnot-push-gatedentries are frozen history: GitHub keeps a workflow's last run for ever, so a workflow that stopped running would otherwise be reported red permanently.stale-greenpassed on an older tree.in-flighthas been running for less than 6h.60df757and head is8a0ec08-- that pass is evidence about a tree nobody runs any more74c6e65and head is8a0ec08-- that pass is evidence about a tree nobody runs any moredad7b04and head is3b40a9a-- that pass is evidence about a tree nobody runs any more6d1f80aand head is3b40a9a-- that pass is evidence about a tree nobody runs any more98588e8and head is3b40a9a-- that pass is evidence about a tree nobody runs any more01e87e5and head is6cfad96-- that pass is evidence about a tree nobody runs any more54311b1and head is21ec989-- that pass is evidence about a tree nobody runs any moreSwept 41 repositories in 217 API reads. https://github.com/OpenAdaptAI/openadapt-ops/actions/runs/33962789306