Let Claude Desktop and Cursor actually watch videos (MCP, fully local)

A developer released claude-real-video, an open-source tool that lets AI agents like Claude Desktop and Cursor watch videos by extracting scene-aware keyframes and timestamped transcripts locally. The tool, available as an MCP server, reduces token usage by deduplicating frames and supports transcription via Whisper.

My AI agent learned to watch videos this week. Here is the 3-line version of how to give yours the same ability. claude-real-video https://github.com/HUANGCHIHHUNGLeo/claude-real-video MIT, 1.9k stars extracts scene-aware, deduplicated keyframes plus a timestamped transcript from any video URL or local file — all processing on your machine. Since 0.8.0 it ships as an MCP server, so MCP clients can request a video directly. pip install 'claude-real-video mcp ' Claude Code: claude mcp add crv -- crv-mcp Claude Desktop — add to claude desktop config.json : { "mcpServers": { "crv": { "command": "crv-mcp" } } } Cursor and any other MCP client: same stdio command, crv-mcp . Two tools show up: watch video source, max frames, language, transcribe — downloads URL or reads local file , extracts keyframes at scene changes, drops near-duplicates, transcribes with Whisper, and returns the transcript plus the first batch of frames as images straight into the conversation. get frames source, start index, count — pages through the rest of the frames.Analyses are cached per source under ~/.cache/crv-mcp , so a follow-up question about the same video is instant. Fixed-interval sampling wastes context on near-identical frames and still misses fast cuts. Scene detection keeps the frames that actually differ: a 58-second test clip goes from 58 sampled frames to 26 that matter. Fewer tokens, nothing missed. The benchmark folder in the repo has the full comparison, reproducible. pip install 'claude-real-video whisper ' or fast ; frames work without it.