Understand and extract insights from video content

TwelveLabsToolkit is a CAMEL toolkit for video understanding powered by TwelveLabs' Pegasus video-language model. Unlike frame-sampling approaches, Pegasus reasons natively over the video - visuals, motion, and audio together - letting an agent answer questions about, summarize, or extract information from a video given only its URL.

Use it in multi-agent tasks where one agent needs to ground its reasoning in actual video content - answering questions about a video, summarizing it, or extracting specific information - without needing to sample and analyze individual frames itself.

Configuration parameters: api_key (optional, read from the TWELVELABS_API_KEY environment variable if not passed directly), model_name (defaults to pegasus1.5), max_tokens (defaults to 2048, capping response length), and timeout (optional, no timeout applied if unset). A free API key with a generous free tier is available at twelvelabs.io.

Integrates as a CAMEL agent toolkit, taking a video URL as input and returning Pegasus's native multimodal analysis (visual, motion, and audio reasoning combined) rather than a frame-by-frame summary assembled after the fact.

Multi-agent system builders who need an agent to reason directly about video content - answering, summarizing, or extracting facts from a video by URL - without building a separate video-processing pipeline.