AI analysis grounded in the code graph — computed facts, not vibes · 2026-07-06T05:40:57Z
video-use is a command-line video-editing toolkit driven by an LLM coding agent (Claude Code, Codex, and similar). Rather than exposing a GUI, it ships a helpers/ module of 51 Python symbols that an agent invokes to transcribe raw footage, cut filler words, colour-grade clips, burn subtitles, and composite a final render. The design point is that the LLM never watches the video — it reads a packed transcript and a rendered timeline view — so the agent can make word-boundary cuts programmatically. It targets developers already working inside an agentic coding environment who want to edit footage from a shell.
What changed recently, how it's actually built (from the code graph), and whether you should care. Free account — no card, no spam.