chore: remove the standalone OCM-98 workflow - #235
Conversation
|
🦞👀 Pull request received. I will update this pull request when review starts. ClawSweeper review completeClawSweeper finished reviewing this revision. The review result is being finalized. |
clawsweeper
Bot
added
P3
|
Codex review: needs maintainer review before merge. Reviewed September 15, 2026, 8:33 PM ET / September 16, 2026, 00:33 UTC (Revision 3). ClawSweeper reviewWhat this changesRemoves the standalone historical runtime-install reproduction workflow while retaining normal CI, isolation tests, and manual reproduction scripts. Merge readiness✅ Ready for maintainer review This remains a useful, narrowly scoped cleanup that current main has not implemented. No actionable defect was found, and the MEMBER-authored PR is exempt from automatic closure. Priority: P3 Review scores
Verification
How this fits togetherOCM manages isolated OpenClaw environments and runtime installations. Its CI runs regression tests, while the removed workflow separately built current and legacy binaries to collect historical installation-isolation evidence. flowchart TD
A[Pull requests and main pushes] --> B[Normal CI]
B --> C[Runtime isolation tests]
D[Manual historical comparison] --> E[Retained reproduction script]
F[Current and legacy binaries] --> E
E --> G[Database and state evidence]
Before mergeNone. Agent review detailsSecurityNone. Review metrics
Technical reviewBest possible solution: Keep runtime-isolation coverage in normal CI and retain the standalone script for deliberate historical comparisons. Do we have a high-confidence way to reproduce the issue? Not applicable: this removes a historical workflow rather than repairing a runtime failure; the deletion and retained coverage are directly inspectable. Is this the best way to solve the issue? Yes. Removing only the redundant workflow is a narrow solution that preserves regression tests and the manual diagnostic path. AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning medium; reviewed against a87053dd608e. LabelsLabel justifications:
EvidenceWhat I checked:
Likely related people:
Rating scale
Overall follows the weaker of proof and patch quality. Workflow
History |
f1be6b2 to
63cc902
Compare
63cc902 to
3811190
Compare
|
Simplified: removed the standalone OCM-98 workflow while preserving normal CI coverage and the diagnostic scripts. |
1 participant
Related: #131.
What Problem This Solves
OCM still carries a dedicated GitHub Actions workflow for the historical runtime-install isolation investigation in #98. It runs manually or on one reproduction branch, repeats formatting and test checks covered by normal CI, builds current and older OCM binaries, and uploads the comparison results.
Keeping that experiment as a separate workflow also gives Dependabot another set of action versions to maintain. The checkout, Rust-cache, and artifact-upload updates in #230, #229, and #227 are confined to this file. The runtime isolation regression already belongs to the normal test suite, while the standalone reproduction script remains available when a historical comparison is useful.
User Impact
This removes the standalone OCM-98 workflow definition from the default branch. Maintainers retain the runtime-isolation tests, the reproduction script and its database helper, and the existing diagnostic evidence. A deliberate reproduction can still use the script with the selected current and legacy OCM binaries.
Normal pull-request CI, release validation, signing and notarization, and npm publication continue through their existing workflows. OCM commands, installed environments, dependency versions, and release behavior are unaffected by this maintenance change.
Why This Change Was Made
The ongoing requirement is to preserve the runtime-isolation repair and its regression coverage. Normal CI already runs the locked Rust test suite containing that regression; maintaining a second GitHub workflow is not needed to keep the repair covered.
The change therefore removes only the standalone workflow declaration. The source tree has no external references that require a replacement or documentation rewrite. The diagnostic script and database helper remain unchanged, and no new workflow, runner, package script, or dependency configuration takes the deleted file's place.
Evidence
Static validation confirms that the complete change is one file deletion. All 214 remaining tracked entries retain their exact Git objects and file modes relative to the current base, including the runtime implementation and tests, both reproduction helpers, every remaining workflow, and the Cargo and Dependabot configuration. A tracked-source reference scan found no dangling workflow filename, workflow name, or dedicated-branch references.
CI on the final commit,
3811190passed all nine jobs, including the full Linux/macOS suites, native installation smoke checks, Windows and minimum-Rust checks, formatting, and npm verification. The retained runtime-install isolation regression explicitly passed in the Linux job and macOS job. All four CodeQL analyses also passed.Independent review found no issues in the complete contribution. Historical diagnostic proof from #131 remains available as historical evidence; removing this workflow did not require rebuilding the old binary or repeating that comparison. No local Rust build, test, or native execution was performed for the static preflight, and no additional reproduction runner was created.