add TwelveLabs video understanding and embedding tool - #254
Open
mohit-twelvelabs wants to merge 2 commits into
Open
add TwelveLabs video understanding and embedding tool#254mohit-twelvelabs wants to merge 2 commits into
mohit-twelvelabs wants to merge 2 commits into
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
✅ Files skipped from review due to trivial changes (1)
🚧 Files skipped from review as they are similar to previous changes (1)
WalkthroughA new ChangesTwelveLabs Video Tool
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 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/concepts/tool_system/twelvelabs_video.md`:
- Around line 79-93: The quickstart examples for TwelveLabsVideo are incorrectly
using the protected _run() method instead of the public run() method provided by
BaseTool. Replace both calls to tool._run() (the Marengo embedding example and
the Pegasus video understanding example) with tool.run() and convert the keyword
arguments into a dictionary passed as the input parameter, ensuring all task
parameters like task, text, video_url, and prompt are properly formatted as
dictionary keys and values.
In
`@omagent-core/src/omagent_core/tool_system/tools/twelvelabs_video/twelvelabs_video.py`:
- Around line 134-147: The embed method in the twelvelabs_video class currently
accepts both text and image_url parameters but silently prioritizes text when
both are provided, creating ambiguous behavior. After the existing validation
check that ensures at least one of text or image_url is provided, add an
additional validation to reject the case where both parameters are provided
simultaneously. Raise a ValueError with a clear message indicating that exactly
one of text or image_url must be provided (not both).
- Around line 74-85: The api_key field in the class has a default value of None,
which bypasses the Pydantic v2 validator when the field is omitted during
instantiation. Remove the default value assignment (= None) from the api_key
field definition to make it required, changing it from Optional[str] = None to
just str. This ensures the api_key_validator will execute and fail at
construction time when the API key is not provided, rather than deferring the
failure to runtime.
🪄 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: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 64da1db1-0ec2-4a6d-b4e7-6cf9d4076b73
📒 Files selected for processing (4)
docs/concepts/tool_system/twelvelabs_video.mdomagent-core/pyproject.tomlomagent-core/src/omagent_core/tool_system/tools/twelvelabs_video/__init__.pyomagent-core/src/omagent_core/tool_system/tools/twelvelabs_video/twelvelabs_video.py
…blic run() in docs
- Make api_key a required field so the validator runs at construction
instead of being bypassed for the default and deferring failure to runtime.
- Reject embed calls that supply both text and image_url for deterministic behavior.
- Update quickstart docs to call the public run({...}) API instead of _run().
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.
Hi! I'm Mohit, I work at TwelveLabs (@mohit-twelvelabs).
What this adds
A new opt-in agent tool,
TwelveLabsVideo, that gives OmAgent's multimodal agents access to the TwelveLabs video AI models:task: "analyze") — video understanding. Answer a natural language prompt about a video by URL (summarize, list objects, describe events, etc.).task: "embed") — multimodal embeddings. Produce a 512-dim embedding vector from text or an image, in a shared text/image/video embedding space (useful for video retrieval).It follows the existing tool convention exactly (
@registry.register_tool()on aBaseToolsubclass with anARGSCHEMA, mirroringTavilyWebSearch), so it is auto-discovered byregistry.import_module()and usable from any tool config.Why it helps
OmAgent already targets multimodal/video agents. This lets an agent reason about videos by URL without standing up a local video model — the heavy lifting runs server-side at TwelveLabs.
Opt-in / non-breaking
Nothing is wired into defaults. The tool only activates when added to a tool config and given an API key:
The
twelvelabsSDK import is lazy (inside__init__), so it is only required when the tool is actually instantiated.How it was tested
embedcall returns a 512-dim vector. The Pegasusanalyzerequest wiring was verified end-to-end (the request reaches and is validated by the backend, which then fetches the video server-side).twelvelabs ^1.2.8inomagent-core/pyproject.toml; docs added atdocs/concepts/tool_system/twelvelabs_video.mdmirroring the Tavily tool doc.You can grab a free API key at https://twelvelabs.io — there's a generous free tier.
Summary by CodeRabbit