Skip to content

fix(ci): restore catalog publishing - #1

Merged
steipete merged 2 commits into
mainfrom
fix/catalog-publish-dependencies
Aug 10, 2026
Merged

steipete merged 2 commits into
mainfrom
fix/catalog-publish-dependencies

Conversation

@steipete

Copy link
Copy Markdown
Contributor

Summary

  • follow the upstream model-catalog publisher's JavaScript-to-TypeScript migration
  • refresh Checkout, Setup Node, and the Node 24 runtime to their current releases
  • regenerate the catalog pricing data that accumulated while scheduled publishing was broken

Root cause

The upstream OpenClaw script migrated from scripts/publish-model-catalog.mjs to scripts/publish-model-catalog.mts. The catalog workflow still called the removed JavaScript entrypoint, so every scheduled run failed after dependency installation with MODULE_NOT_FOUND.

Verification

  • actionlint -color
  • generated a valid 40-provider, 232-model catalog locally on Node 24.19.0 using the current upstream checkout and live pricing sources
  • upstream focused publisher tests: 8 passed
  • local HTTP artifact fetch: 200, schema v1, 40 providers, 232 models
  • hosted branch workflow: https://github.com/openclaw/catalog/actions/runs/31362044455
  • structured autoreview: clean, no accepted/actionable findings

@clawsweeper

clawsweeper Bot commented Aug 10, 2026

Copy link
Copy Markdown

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

@steipete
steipete merged commit 62da41c into main Aug 10, 2026
2 checks passed
@steipete
steipete deleted the fix/catalog-publish-dependencies branch August 10, 2026 06:32
@clawsweeper clawsweeper Bot added merge-risk: 🚨 automation 🚨 Merging this PR could break CI, automerge, proof capture, label sync, or automation. P1 Urgent regression or broken agent/channel workflow affecting real users now. proof: sufficient Contributor real behavior proof is sufficient. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. labels Aug 10, 2026
@clawsweeper

clawsweeper Bot commented Aug 10, 2026

Copy link
Copy Markdown

Codex review: needs maintainer review before merge. Reviewed August 10, 2026, 2:31 AM ET / 06:31 UTC.

ClawSweeper review

What this changes

The PR switches scheduled catalog publishing to the upstream TypeScript publisher, updates pinned GitHub Actions and Node 24, and refreshes the generated pricing catalog.

Merge readiness

⚠️ Ready for maintainer review - 2 items remain

Keep open for merge review: current main still invokes the obsolete JavaScript publisher entrypoint, while this PR makes the narrow workflow repair and refreshes the resulting catalog data.

Priority: P1
Reviewed head: 75b39eaaa7481d6673cf2042d0eade1b1768e523

Review scores

Measure Result What it means
Overall readiness 🐚 platinum hermit (4/6) A focused repair with credible real-run evidence and no actionable correctness or security finding.
Proof confidence 🐚 platinum hermit (4/6) ✨ media proof bonus Sufficient (linked_artifact): The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.
Patch quality 🦞 diamond lobster (5/6) No actionable review findings were identified.

Verification

Check Result Evidence
Real behavior Verified Sufficient (linked_artifact): The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.
Evidence reviewed 4 items Current main retains the failing invocation: The scheduled publisher on current main still runs the JavaScript entrypoint at line 44; the PR changes that command to the TypeScript entrypoint with the installed tsx loader.
Narrow repair and regenerated output: The first PR commit changes only five workflow lines; its follow-up updates the generated catalog by 114 additions and 108 removals.
After-fix runtime evidence: The PR body reports a Node 24.19.0 generation, a schema-v1 local artifact fetch with 40 providers and 232 models, focused upstream tests, and a hosted branch workflow run.
Findings None None.
Security None None.

How this fits together

A scheduled GitHub Actions job checks out OpenClaw, generates the hosted model catalog, and commits it when model metadata or pricing changes. OpenClaw installations periodically fetch that catalog to refresh model information.

flowchart LR
  A[Scheduled trigger] --> B[Checkout catalog]
  B --> C[Checkout OpenClaw]
  C --> D[Install dependencies]
  D --> E[TypeScript catalog publisher]
  E --> F[Generated catalog JSON]
  F --> G[Commit changed catalog]
  G --> H[OpenClaw model refresh]
Loading

Before merge

  • Resolve merge risk (P1) - The job intentionally follows OpenClaw main, so a future upstream publisher or runtime migration can again require a compatible workflow adjustment despite this validated repair.
  • Complete next step (P2) - This is a complete contributor patch awaiting ordinary maintainer merge review, not a separate bounded repair-lane task.
Agent review details

Security

None.

Review metrics

Metric Value Why it matters
Workflow repair 5 changed lines The functional repair is confined to the scheduled publisher invocation and its pinned runtime actions.
Catalog refresh 114 added, 108 removed The generated snapshot carries the recovered pricing and model metadata update.

Merge-risk options

Maintainer options:

  1. Land and observe the next schedule (recommended)
    Merge the validated repair and confirm the next four-hour publishing run successfully commits an update or skips unchanged content.

Technical review

Best possible solution:

Merge the tested TypeScript publisher update and refreshed catalog, then confirm the next scheduled run publishes or cleanly skips unchanged content.

Do we have a high-confidence way to reproduce the issue?

Yes—source inspection gives a high-confidence path: current main still invokes the JavaScript publisher at workflow line 44, while the reported upstream entrypoint is TypeScript. This read-only review did not execute an external OpenClaw checkout itself.

Is this the best way to solve the issue?

Yes. Updating the existing scheduled command to the current upstream entrypoint preserves the established publishing and commit-on-change flow without adding a parallel mechanism.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against aa1e23af1498.

Labels

Label changes:

  • add P1: A broken scheduled catalog publisher leaves model metadata and pricing stale for installations that refresh the hosted catalog.
  • add merge-risk: 🚨 automation: The PR changes the production scheduled publishing workflow and therefore needs post-merge schedule observation beyond ordinary static review.
  • add proof: sufficient: Contributor real behavior proof is sufficient. The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.
  • add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🦞 diamond lobster.
  • add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (linked_artifact): The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.

Label justifications:

  • P1: A broken scheduled catalog publisher leaves model metadata and pricing stale for installations that refresh the hosted catalog.
  • merge-risk: 🚨 automation: The PR changes the production scheduled publishing workflow and therefore needs post-merge schedule observation beyond ordinary static review.
  • rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🦞 diamond lobster.
  • status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (linked_artifact): The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.
  • proof: sufficient: Contributor real behavior proof is sufficient. The PR supplies after-fix real-run evidence for generation, schema-serving, focused upstream tests, and a hosted branch workflow; its linked workflow is the relevant production-path trace.

Evidence

What I checked:

  • Current main retains the failing invocation: The scheduled publisher on current main still runs the JavaScript entrypoint at line 44; the PR changes that command to the TypeScript entrypoint with the installed tsx loader. (.github/workflows/publish.yml:44, aa1e23af1498)
  • Narrow repair and regenerated output: The first PR commit changes only five workflow lines; its follow-up updates the generated catalog by 114 additions and 108 removals. (.github/workflows/publish.yml:21, 25f7d7e31d4f)
  • After-fix runtime evidence: The PR body reports a Node 24.19.0 generation, a schema-v1 local artifact fetch with 40 providers and 232 models, focused upstream tests, and a hosted branch workflow run. (models/v1/catalog.json:1, 75b39eaaa748)
  • Workflow provenance: Blame attributes the current scheduled publishing workflow, including its original publisher command, to the catalog automation commit. (.github/workflows/publish.yml:44, 099f938030cc)

Likely related people:

  • openclaw-catalog[bot]: Git blame assigns the current publishing workflow to its catalog automation commit, and current-main catalog updates are authored by the same automation identity. (role: workflow originator and recent catalog updater; confidence: medium; commits: 099f938030cc, aa1e23af1498; files: .github/workflows/publish.yml, models/v1/catalog.json)

Rating scale

Score Internal tier Crab rank Meaning
6/6 S 🦀 challenger crab Exceptional readiness
5/6 A 🦞 diamond lobster Very strong readiness
4/6 B 🐚 platinum hermit Good normal PR; ordinary maintainer review
3/6 C 🦐 gold shrimp Useful, but confidence is limited
2/6 D 🦪 silver shellfish Proof or implementation needs work
1/6 F 🧂 unranked krab Not merge-ready
N/A NA 🌊 off-meta tidepool Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

  • ClawSweeper keeps one durable marker-backed review comment per issue or PR.
  • Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
  • A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
  • PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
  • Maintainers can also comment @clawsweeper review to request a fresh review only.
  • Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
  • Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
  • Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

merge-risk: 🚨 automation 🚨 Merging this PR could break CI, automerge, proof capture, label sync, or automation. P1 Urgent regression or broken agent/channel workflow affecting real users now. proof: sufficient Contributor real behavior proof is sufficient. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant