Open
Conversation
content_to_text() was stripping image_url parts from ToolMessage content before sending to the API. This meant VLM models never saw screenshots returned by browser tools (CUA mode). Use normalize_content() instead, consistent with how all other message types handle multipart content.
Collaborator
Author
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
`content_to_text()` in `OpenAIChatCompletionsClient.to_native_prompt` was stripping `image_url` parts from `ToolMessage` content before sending to the API. This meant VLM models never saw screenshots returned by browser tools (CUA mode).
Uses `normalize_content()` instead, consistent with how system, user, and assistant messages already handle multipart content.
Note
Low Risk
Low risk: a small change to prompt serialization so
ToolMessagecontent is no longer flattened to text, but it could alter what gets sent to the OpenAI API for tool responses.Overview
Ensures
ToolMessagecontent passed to the Chat Completions API preserves multipart payloads (e.g.,image_urlparts) by switching fromcontent_to_text()tonormalize_content()inOpenAIChatCompletionsClient.to_native_prompt.This aligns tool-message serialization with system/user/assistant handling so non-text tool outputs are no longer dropped before reaching vision-capable models.
Written by Cursor Bugbot for commit 2404c42. This will update automatically on new commits. Configure here.