A new TwelveLabsTools toolkit lets your agents analyze videos and generate multimodal text embeddings, so video becomes something an agent can search, summarize, and reason over rather than a black box. Drop it in the way you would any other Agno toolkit.
See the TwelveLabs docs for setup.