molecule-core/.github/workflows
Hongming Wang 9d4ab7b1a2 feat(ci): auto-promote-on-e2e — retag :latest on green E2E Staging SaaS
Closes the final gap in the SaaS pipeline. After auto-promote-staging
fast-forwards main, publish-workspace-server-image builds new
`:staging-<sha>` images, but `:latest` (what prod tenants pull) only
moves on either a manual `promote-latest.yml` dispatch or a canary-
verify retag (gated on Phase 2 fleet that doesn't exist).

This workflow closes that gap by retagging
`platform:staging-<sha>` + `platform-tenant:staging-<sha>` → `:latest`
whenever E2E Staging SaaS passes for a `main` push. Uses crane
(no Docker daemon needed). Verifies both images exist before retagging
either, so a half-published state is impossible.

Why trigger only on `main` (not staging):
  - `:latest` is what prod tenants pull. Only SHAs that have reached
    `main` (via auto-promote-staging) should advance `:latest`.
  - Triggering on staging would let a staging-only revert advance
    `:latest` to a SHA that never reaches `main`, breaking the
    invariant "production runs what's on `main`".

Why a separate workflow rather than folding into e2e-staging-saas.yml:
  - Test concerns and release concerns separate.
  - Disabling promote during an incident is one workflow toggle, not
    an edit to the long E2E file.
  - When Phase 2 canary work eventually lands, the canary path can
    replace this trigger without touching the E2E workflow.

Doc-aligned: per molecule-controlplane/docs/canary-tenants.md,
"green staging E2E → :latest" is the recommended approach for the
current scale (≤20 paying tenants); canary fleet is deferred until
blast radius grows.

Pipeline after this lands is fully self-healing:
  staging push → 4 gates green → auto-promote fast-forwards main
   → publish-workspace-server-image → E2E Staging SaaS
   → THIS WORKFLOW retags :latest → tenant fleet auto-pulls in 5 min
                                    (or redeploy-tenants-on-main fans out faster)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-28 13:58:41 -07:00
..
auto-promote-on-e2e.yml feat(ci): auto-promote-on-e2e — retag :latest on green E2E Staging SaaS 2026-04-28 13:58:41 -07:00
auto-promote-staging.yml fix(ci): auto-promote gate-check uses workflow file paths, not names 2026-04-28 13:15:13 -07:00
auto-tag-runtime.yml feat(platform/admin): /admin/workspace-images/refresh + Docker SDK + GHCR auth 2026-04-26 10:17:21 -07:00
block-internal-paths.yml fix(ci): block-internal-paths handle merge_group + shallow-clone BASE 2026-04-26 23:54:00 -07:00
canary-staging.yml Merge branch 'staging' into ci/cicd-review-quick-wins 2026-04-27 13:27:16 -07:00
canary-verify.yml ci: canary-verify graceful-skip + draft auto-promote staging→main 2026-04-22 22:39:23 +00:00
check-merge-group-trigger.yml ci: add linter that fails when required workflow lacks merge_group trigger 2026-04-24 00:33:05 -07:00
ci.yml test(workspace): centralize pytest-cov config + 92% floor (closes #1817) 2026-04-26 06:21:22 -07:00
codeql.yml ci: add merge_group trigger to ci + codeql 2026-04-23 21:24:53 -07:00
e2e-api.yml fix(ci): e2e gates always emit a result so auto-promote can read it 2026-04-28 12:43:26 -07:00
e2e-staging-canvas.yml fix(ci): e2e gates always emit a result so auto-promote can read it 2026-04-28 12:43:26 -07:00
e2e-staging-saas.yml Merge branch 'staging' into ci/cicd-review-quick-wins 2026-04-27 13:27:16 -07:00
e2e-staging-sanity.yml fix(e2e): CP DELETE /cp/admin/tenants body uses 'confirm', not 'confirm_token' 2026-04-21 04:50:28 -07:00
pr-guards.yml ci: add pr-guards caller that disables auto-merge on push 2026-04-27 06:39:31 -07:00
promote-latest.yml perf(ci): move all public-repo workflows to ubuntu-latest 2026-04-22 12:56:49 -07:00
publish-canvas-image.yml perf(ci): move all public-repo workflows to ubuntu-latest 2026-04-22 12:56:49 -07:00
publish-runtime.yml feat(cascade): verify wheel content sha256 against just-built dist 2026-04-28 10:53:50 -07:00
publish-workspace-server-image.yml ci(publish-image): also tag :staging-latest so CP auto-picks up new builds 2026-04-24 00:29:55 -07:00
redeploy-tenants-on-main.yml ci(redeploy): fire post-main tenant fleet redeploy via CP admin endpoint 2026-04-24 14:34:28 -07:00
retarget-main-to-staging.yml ci(retarget): handle 422 'duplicate PR' by closing redundant main-PR (closes #1884) 2026-04-26 00:53:55 -07:00
runtime-pin-compat.yml ci(pin-compat): split into two workflows so each gets a narrow paths filter 2026-04-28 10:50:09 -07:00
runtime-prbuild-compat.yml ci(pin-compat): split into two workflows so each gets a narrow paths filter 2026-04-28 10:50:09 -07:00
secret-scan.yml chore(secret-scan): add sk-cp- MiniMax pattern (F1088 retroactive fix) 2026-04-26 21:43:22 -07:00
sweep-cf-orphans.yml fix(ci): stop sweep-cf-orphans noise — drop merge_group + soft-skip when secrets unset 2026-04-26 08:05:53 -07:00
sweep-stale-e2e-orgs.yml ci: hourly sweep of stale e2e-* orgs on staging 2026-04-24 23:07:57 -07:00
test-ops-scripts.yml refactor(ops): apply simplify findings on #2027 PR 2026-04-26 00:28:15 -07:00