🤖 Add image reading support to file_read tool #357
Conversation
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
- Detect MIME types using mime-types package - Read images as base64 binary data - Use toModelOutput to send images as media content to AI models - Update UI to display image previews - Add comprehensive tests for image reading and conversion
This is the critical missing piece - without passing tools to convertToModelMessages, the toModelOutput function is never called and images are sent as JSON instead of media content. Added test to verify image tool results are converted to media content when tools are provided to convertToModelMessages.
- Use async fs.promises.writeFile instead of sync - Use optional chaining for toolMessage checks - Add eslint-disable for unavoidable any assignment
kylecarbs
force-pushed
the
allow-reading-images
branch
from
58c4326 to
c728067
Compare
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Outdated suggestions cannot be applied.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.
The file_read tool now properly reads images and sends them to AI models as media content instead of trying to parse them as text.
Implementation:
Added
toModelOutputfunction to file_read tool that detects images by MIME type and returns them in the AI SDK's media content format. The AI SDK'sconvertToModelMessagesautomatically uses this when converting tool results.Changes:
toModelOutputto convert images to media content blocksTesting:
Added tests for PNG/JPEG reading, MIME type detection, and
toModelOutputconversion. All 18 file_read tests pass.Generated with
cmux