Skip to content

feat(groom): default the agent model to claude-opus-5#70

Open
mattmillerai wants to merge 1 commit into
mainfrom
matt/groom-model-opus-5
Open

feat(groom): default the agent model to claude-opus-5#70
mattmillerai wants to merge 1 commit into
mainfrom
matt/groom-model-opus-5

Conversation

@mattmillerai

Copy link
Copy Markdown
Contributor

Bumps the groom model input default claude-opus-4-8claude-opus-5, and documents the trade-off in the input description.

The finder, verifier, and builder agents are groom's output quality — surfacing real refactors, adjudicating CONFIRM/DOWNGRADE/REJECT, and authoring the patch — so the default should track the current flagship. Callers that want to hold a run steady can still pin an older id via the model: input; neither current caller (comfy-cloud-mcp-server, cloud) overrides it, so both pick this up on their next SHA bump.

ELI5

Groom uses an AI model to read the repo, decide which cleanups are real, and write the fix. This points it at the newest, strongest model instead of the previous one, because the quality of what it finds and writes depends almost entirely on that choice. Anyone who wants the old model can still ask for it by name.

Cost note: a finder pass is ~$7.50 at the previous model; expect this to shift with Opus 5 pricing. Runtime is unchanged (~12 min, well inside the 40-min timeout).

actionlint clean.

The finder/verifier/builder agents drive groom's output quality — finding real
refactors, adjudicating CONFIRM/REJECT, and writing the patch — so the default
tracks the current flagship. Callers can still pin an older id (e.g.
claude-opus-4-8) via the model input to hold a run steady.

actionlint clean.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@coderabbitai

coderabbitai Bot commented Jul 24, 2026

Copy link
Copy Markdown

Review Change Stack

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 1123244a-7709-4b60-a4e4-3907943b5292

📥 Commits

Reviewing files that changed from the base of the PR and between 29a81ca and 70242b4.

📒 Files selected for processing (1)
  • .github/workflows/groom.yml

📝 Walkthrough

Walkthrough

The reusable groom workflow’s model input now documents the finder, verifier, and builder agents, with its default changed from claude-opus-4-8 to claude-opus-5.

Changes

Groom workflow model input

Layer / File(s) Summary
Update model input contract
.github/workflows/groom.yml
The workflow_call.inputs.model description includes the builder agent, and its default model changes to claude-opus-5.
🚥 Pre-merge checks | ✅ 2
✅ Passed checks (2 passed)
Check name Status Explanation
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Commit unit tests in branch matt/groom-model-opus-5
✨ Simplify code
  • Create PR with simplified code
  • Commit simplified code in branch matt/groom-model-opus-5

Comment @coderabbitai help to get the list of available commands.

@mattmillerai mattmillerai added the cursor-review Multi-model cursor review label Jul 24, 2026

@github-actions github-actions Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🔍 Cursor Review — Consolidated panel

Triggered by @mattmillerai.

✅ No high-signal findings.

Panel: 8/8 reviewers contributed findings.

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

Labels

cursor-review Multi-model cursor review

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant