fix(look-at): respect configured multimodal-looker model instead of overriding via dynamic fallback#2686
Merged
code-yeongyu merged 1 commit intocode-yeongyu:devfrom Mar 25, 2026
Conversation
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
There was a problem hiding this comment.
No issues found across 3 files
Confidence score: 5/5
- Automated review surfaced no issues in the provided summaries.
- No files require special attention.
Auto-approved: Fixes a bug where valid vision models were incorrectly overridden or blocked due to an incomplete metadata cache. Simplifies logic to correctly prioritize user-configured models.
…verriding via dynamic fallback
22dd884 to
db32bad
Compare
sjawhar
added a commit
to sjawhar/oh-my-opencode
that referenced
this pull request
Mar 26, 2026
Rebased onto dev@upstream (v3.13.1). Dropped fix/look-at-respect-configured-model and fix/atlas-subagent-agent-guard (merged upstream as PRs code-yeongyu#2686 and code-yeongyu#2751).
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
The
look_attool's model resolution (resolveMultimodalLookerAgentMetadata) was silently overriding the user's configuredmultimodal-lookermodel with a dynamically resolved one from the vision-capable models cache. This caused failures when the dynamically selected model (e.g., an Antigravity-prefixed Gemini model) was incompatible with the provider's API.Problem
When a user configures
multimodal-lookerwith a specific model (e.g.,openai/gpt-5.4), the resolution function:provider.*.modelswithmodalities.input: ["image"])This was confirmed via logs: the tool resolved to
google/antigravity-gemini-3-proinstead of the configuredopenai/gpt-5.4, causing prompt failures.Fix
multimodal-looker, use it regardless of the vision-capable cache. The cache is incomplete (only includes models with explicitmodalities.inputdeclarations in the provider config).isVisionCapableResolvedModelgate intools.tsthat would block the configured model for the same reason.Testing
Summary by cubic
Ensures
look_atuses the configuredmultimodal-lookermodel and variant, only falling back to dynamic resolution when none is set. Prevents silent overrides to incompatible vision models and the resulting prompt failures.tools.ts.Written for commit db32bad. Summary will update on new commits.