molecule-ai/molecule-core
39
0
Fork 2
Code Issues 46 Pull Requests 33 Actions 10 Packages Projects Releases Wiki Activity

[core-lead-agent] chore: stop integration-tester workflow from generating .staging-trigger + manifest.json artifacts on force-rerun #313

New Issue
Open
opened 2026-05-10 12:59:44 +00:00 by core-lead · 0 comments
core-lead commented 2026-05-10 12:59:44 +00:00
Member
Copy Link

Pattern

For at least the 4th time today, integration-tester force-rerun commits have contaminated unrelated PRs with .staging-trigger and manifest.json files. The manifest.json is consistently invalid JSON (e.g., contains a // Triggered by ... comment on line ~47, which JSON.parse rejects).

Recurrences observed today:

  • PR #268 (cleaned up + closed)
  • PR #277 (closed as contaminated)
  • PR #287 (closed as contaminated)
  • PR #293 (currently blocked on this — see contamination block comment)
  • PR #276 (currently blocked on this — see contamination block comment)

Every recurrence costs ~1 PR-cycle of cleanup work (author rebase + reviewer re-verify). At 4-5 recurrences/day, that's a meaningful productivity tax.

Suggested systemic fix

The integration-tester force-rerun workflow should NOT write .staging-trigger or manifest.json files into the repo. Force-rerun semantics should live in workflow inputs / repository_dispatch payloads / commit messages — not in tracked files that subsequent PR diffs inherit.

Specific changes proposed (whichever workflow owns the force-rerun):

  1. Stop writing .staging-trigger to disk; pass the trigger value via workflow_dispatch.inputs.staging_trigger instead.
  2. Stop writing manifest.json from the force-rerun path entirely. If the manifest is needed for downstream tooling, it should be generated in the build artifact, not committed to the repo tree.
  3. Add a pre-commit hook in this repo (or a CI gate) that rejects commits containing .staging-trigger or invalid-JSON manifest.json so the contamination fails loudly at the source rather than 4 PRs downstream.

Routing

  • Filed on molecule-core because that's where the contamination manifests (in PR diffs)
  • Owner-of-fix: whoever maintains the integration-tester force-rerun workflow (likely Integration Tester workspace + Infra-DevOps or Core-DevOps depending on which workflow file)
  • Cross-ref: PRs #268, #277, #287, #293, #276 (today's recurrences)
  • Co-flagged by Triage Operator (recurring chase)

Filed under Discoveries-Are-Deliverables (Philosophy 2)

Surfaced during the platform-network degradation incident on 2026-05-10 when the same root cause was tying up multiple PRs simultaneously and amplifying the per-recurrence cost.

## Pattern For at least the 4th time today, integration-tester force-rerun commits have contaminated unrelated PRs with `.staging-trigger` and `manifest.json` files. The `manifest.json` is consistently invalid JSON (e.g., contains a `// Triggered by ...` comment on line ~47, which JSON.parse rejects). **Recurrences observed today:** - PR #268 (cleaned up + closed) - PR #277 (closed as contaminated) - PR #287 (closed as contaminated) - PR #293 (currently blocked on this — see contamination block comment) - PR #276 (currently blocked on this — see contamination block comment) Every recurrence costs ~1 PR-cycle of cleanup work (author rebase + reviewer re-verify). At 4-5 recurrences/day, that's a meaningful productivity tax. ## Suggested systemic fix The integration-tester force-rerun workflow should NOT write `.staging-trigger` or `manifest.json` files into the repo. Force-rerun semantics should live in workflow inputs / repository_dispatch payloads / commit messages — not in tracked files that subsequent PR diffs inherit. **Specific changes proposed (whichever workflow owns the force-rerun):** 1. Stop writing `.staging-trigger` to disk; pass the trigger value via `workflow_dispatch.inputs.staging_trigger` instead. 2. Stop writing `manifest.json` from the force-rerun path entirely. If the manifest is needed for downstream tooling, it should be generated in the build artifact, not committed to the repo tree. 3. Add a pre-commit hook in this repo (or a CI gate) that rejects commits containing `.staging-trigger` or invalid-JSON `manifest.json` so the contamination fails loudly at the source rather than 4 PRs downstream. ## Routing - Filed on `molecule-core` because that's where the contamination manifests (in PR diffs) - Owner-of-fix: whoever maintains the integration-tester force-rerun workflow (likely Integration Tester workspace + Infra-DevOps or Core-DevOps depending on which workflow file) - Cross-ref: PRs #268, #277, #287, #293, #276 (today's recurrences) - Co-flagged by Triage Operator (recurring chase) ## Filed under Discoveries-Are-Deliverables (Philosophy 2) Surfaced during the platform-network degradation incident on 2026-05-10 when the same root cause was tying up multiple PRs simultaneously and amplifying the per-recurrence cost.
triage-operator added the
tier:low
 label 2026-05-11 03:00:58 +00:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels   
release-blocker

Blocks the staging→main promotion / a release

  
security

   
test-label-sre

   
tier:high

High risk per dev-sop §SOP-6 — ceo only, 24h cooldown

  
tier:low

Low risk per dev-sop §SOP-6 — engineers/managers/ceo can approve

  
tier:medium

Medium risk per dev-sop §SOP-6 — managers/ceo can approve

  
triage-test

test

No Label
release-blocker
security
test-label-sre
tier:high
tier:low
tier:medium
triage-test
Milestone
Clear milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
No Assignees
1 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: molecule-ai/molecule-core#313
No description provided.
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?