Developer Builds Stateful AI Video Editor Inside Claude Code Using Google Gemini API
A developer has released an open-source tool called omni-skill-claude that integrates Google's Gemini Omni Flash video model into Claude Code via a lightweight MCP server. The project allows users to generate, animate, and edit videos entirely from a terminal by typing natural language prompts without leaving their coding environment. Unlike most video-generation tools, it leverages Google's stateful Interactions API, meaning the model retains visual context across multiple editing turns so users can refine clips incrementally without re-describing entire scenes. The skill supports five workflows including text-to-video, image animation, keyframe interpolation, subject-consistent generation, and video restyling. Completed videos can also be uploaded directly to YouTube through the same interface.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in