Refresh task agent model routing#65
Merged
Merged
Conversation
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
🔗 Linked repositories identifiedCodeRabbit considers these linked repositories for cross-repo context during reviews:
Summary
Validation
WalkthroughAdd independent audit routing and triage effort configuration, update model defaults and write-probe mappings, wire dedicated agent options, expose routing metadata, and synchronize tests, user documentation, architecture notes, and supervisor guidance. ChangesWorkflow routing
Sequence Diagram(s)sequenceDiagram
participant makeConfig
participant ODWWorkflow
participant AuditAgent
participant TriageAgent
makeConfig->>ODWWorkflow: provide audit and triage configuration
ODWWorkflow->>AuditAgent: invoke audit with adapter, model, and effort
ODWWorkflow->>TriageAgent: invoke triage with adapter, model, and effort
ODWWorkflow->>ODWWorkflow: publish audit and triage modelRouting
Suggested reviewers: Poem
🚥 Pre-merge checks | ✅ 19 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (19 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
leynos
marked this pull request as ready for review
July 19, 2026 09:32
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: dbe63d377c
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Collaborator
|
@coderabbitai review |
✅ Action performedReview finished.
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/architecture.md`:
- Around line 204-208: Update the sentence in the architecture documentation by
inserting a comma after “same Sol model” before “so every triage task,” without
changing the surrounding wording.
In `@docs/users-guide.md`:
- Around line 576-577: Update the assessmentAdapter and assessmentModel
documentation to state that assessmentAdapter defaults to the review adapter,
while assessmentModel independently defaults to claude-sonnet-5. Remove the
wording that implies both share the review adapter and model defaults.
- Around line 579-581: Update the sentence in the documentation so the list
reads “Build, implementation, and integration,” adding the Oxford comma while
preserving the surrounding wording.
In `@skills/df12-build-supervisor/SKILL.md`:
- Around line 176-179: Update the configuration list in SKILL.md to include
triageAdapter, triageModel, triageEffort, and triageEscalationModel. Document
that triage defaults to the codex adapter with gpt-5.6-sol at medium effort,
while its escalation model remains independently configurable; keep the wording
consistent with the referenced architecture, user guide, and config definitions.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: d59ac4f1-733f-43ac-8528-c54c7ebb3385
📒 Files selected for processing (9)
docs/architecture.mddocs/developers-guide.mddocs/users-guide.mdskills/df12-build-supervisor/SKILL.mdsrc/workflows/df12-build-odw/config.tssrc/workflows/df12-build-odw/main.tstests/df12-build-odw-assessment.test.mjstests/modules/config.test.tsworkflows/df12-build-odw.js
🔗 Linked repositories identified
CodeRabbit considers these linked repositories for cross-repo context during reviews:
leynos/nixie(auto-detected)
added 2 commits
July 23, 2026 15:09
Route build work through GPT-5.6 Terra and remediation triage through GPT-5.6 Sol at medium effort. Use Luna and Haiku for minimal-effort write probes, and give post-merge audit its own Sonnet 5 medium route. Keep each route operator-configurable and document the new defaults.
Warn operators upgrading Codex-only sidecars that audit no longer follows the review adapter and can independently require Claude authentication. List the audit routing arguments and summarize the current Sonnet 5 medium default in the supervisor guidance.
lodyai
Bot
force-pushed
the
agent-refresh
branch
from
July 23, 2026 13:11
407c984 to
a5b889b
Compare
Exercise default and overridden adapter, model, and effort values through the shipped workflow artefact. Reuse the same routing helper for terminal metadata so the reported configuration stays covered alongside agent calls. Align the architecture, user guide, and supervisor guidance with the live assessment and triage defaults.
Collaborator
|
@coderabbitai review |
✅ Action performedReview finished.
|
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/users-guide.md`:
- Around line 573-575: Update the triageAdapter and triageModel documentation
entry to list only the adapter and model defaults, codex and gpt-5.6-sol. Remove
the medium effort default from this entry and leave it documented under the
separate triageEffort configuration entry.
In `@skills/df12-build-supervisor/SKILL.md`:
- Around line 138-146: Add triageAdapter to the stale adapter override checklist
in the workflow-copy and relaunch guidance, alongside planAdapter,
reviewAdapter, auditAdapter, and assessmentAdapter. Ensure the instructions
explicitly require auditing and validating triageAdapter against odw.config.json
or ODW’s built-in adapter set before relaunch.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 6ed1995a-40f8-43af-bf0b-5a38a462b4e5
📒 Files selected for processing (10)
docs/architecture.mddocs/developers-guide.mddocs/users-guide.mdskills/df12-build-supervisor/SKILL.mdsrc/workflows/df12-build-odw/config.tssrc/workflows/df12-build-odw/main.tstests/df12-build-odw-assessment.test.mjstests/df12-build-odw-audit-triage-worktree.test.mjstests/modules/config.test.tsworkflows/df12-build-odw.js
🔗 Linked repositories identified
CodeRabbit considers these linked repositories for cross-repo context during reviews:
leynos/nixie(auto-detected)
Owner
Author
|
@coderabbitai resume |
✅ Action performedReviews resumed. |
Separate the triage effort default from the adapter and model entry. Require sidecar relaunch checks to validate stale triage adapter overrides.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This branch right-sizes df12-build's task-agent routing for the refreshed model
portfolio. Build-side work now uses GPT-5.6 Terra at medium effort, remediation
triage uses GPT-5.6 Sol at medium effort, write probes use GPT-5.6 Luna and
Claude Haiku at minimal effort, and post-merge audit has a dedicated Claude
Sonnet 5 medium route.
The routes remain independently configurable, while the defaults, generated
workflow artefact, tests, and operator documentation stay aligned.
Review walkthrough
config.ts
for the new model, adapter, and effort defaults.
main.ts
for the dedicated audit route, explicit triage effort wiring, and reported
model routing.
config.test.ts,
audit and triage workflow tests,
and
df12-build-odw-assessment.test.mjs
for the default, override, effort-wiring, and authentication-preflight
contracts.
ODW workflow artefact
and the updated
architecture,
developers' guide, and
users' guide.
Validation
make all: passedgit diff --check: passedReferences