{"slug": "show-hn-openedit-fast-open-source-video-editing-agent", "title": "Show HN: OpenEdit – fast, open Source video editing agent", "summary": "VEED released OpenEdit, an open-source, agent-driven video editing pipeline that uses VEED's HTML renderer, supporting Claude Code, Codex, and Gemini, with no GUI or timeline. The renderer is up to 2.2x faster than Chrome-driven renderers in internal benchmarks, and the tool is available for Apple Silicon Macs running macOS Tahoe 26.0, with Windows and Linux support planned. OpenEdit allows users to edit, cut, reframe, add subtitles, and create motion graphics through natural language prompts, integrating with services like Fal and Figma MCP.", "body_md": "**Not the editor you rent, but the one you own.**\n\n[Install](#installation) ·\n[Usage](#usage) ·\n[Examples](#examples) ·\n[VEED](https://www.veed.io)\n\nOpenEdit is an open-source, agent-driven editing pipeline that ships with VEED's HTML renderer — closed source, but free to use.\n\nThere is no GUI and no timeline. The pipeline is driven entirely through your coding agent — edit, cut, and reframe footage; layer motion graphics and visual elements; turn slides or websites into video; capture web pages; and pull in any video/image generating service or MCP server when it helps. Source video is optional: stills, slides, generated media, or pure motion graphics are enough when the brief calls for it. Be creative and fluid — try new ideas rather than collapsing every ask onto captions.\n\n*This launch video was made in OpenEdit — click it to watch with sound.*\n\n| Platform | Apple Silicon Mac, macOS Tahoe 26.0 |\n| Intel Macs, earlier macOS | Not supported |\n| Windows, Linux | Planned; prioritisation depends on demand |\n| Transcription | needed only to caption speech — OpenEdit asks once and remembers: VEED (\n|\n\nDownload this repo, or install via the command line with:\n\n```\nnpx skills add veedstudio/open-edit\n```\n\nOpenEdit uses an agent-agnostic skill and repository guide. It supports Claude Code, Codex, and Gemini;\nthe installed skill prepares the runtime and loads its `AGENTS.md`\n\ninstructions explicitly.\n\nOpen your coding agent. For example, with Claude Code:\n\n```\nclaude\n```\n\nthen ask it:\n\n```\nAdd subtitles to my video [VIDEO]\n```\n\nThe agent will automatically transcribe your video, analyse the transcript, create a suitable design for your video, and then render it.\n\nAfter 1-3 minutes, the agent will return your video with subtitles burned in, as well as launch a video previewer. It will also tell you where to find the final MP4.\n\nYou can then ask the agent for subtitle amendments, for example:\n\n```\nI don't like the yellow colour, make it darker\n\nMove the text up a bit\n\nWhen he says \"go buy it now\", make sure the 'now' really stands out.\n```\n\nYou can also hand the agent a reference image to copy a style from:\n\n```\nAdd subtitles that look like this [IMG-REF], to my video [VIDEO]\n```\n\nReal outputs, each with the prompt that produced it. Click any example to watch it with sound.\n\n```\ncreate viral subtitles with /open-edit and translate my video to 5 languages using VEED Lipsync 2.0 on Fal\n```\n\nOne source clip, three languages, three caption styles:\n\n| Spanish | French | German |\n|---|---|---|\n\n```\ngenerate 3 viral hooks in Seedance 2.0 on Fal and create dynamic motion graphics using /open-edit\n```\n\nThree AI-generated hooks, three motion-graphic treatments:\n\n```\nuse Figma MCP to study my BrandBook and create branded campaign graphics using /open-edit\n```\n\nOne brand book, three campaign cards:\n\nCaption styles are authored in HTML and CSS. Rendering does not use a headless browser: there is no browser to install, launch, or keep alive for the duration of a render.\n\nInternal benchmarks measured the renderer up to 2.2x faster than Chrome-driven renderers. These measurements are preliminary and not yet reproducible outside VEED; a documented benchmark and its methodology will follow.\n\nThe renderer is integrated but not required. Composition does not depend on it, so Chrome can be used as the render backend instead.\n\nOpenEdit asks once which provider to use and records the answer at the runtime root\n(`.open-edit-prefs.json`\n\n); it does not ask again. Every provider writes the same file —\n`runs/<key>/transcript.json`\n\n, with real per-word timings — and nothing downstream can tell which one ran.\n\n**VEED** — best quality, and the default workflow. Requires a VEED account; usage limits are your\naccount's, and a free account covers about 2 minutes of transcription a month, beyond which it needs a\n[plan](https://www.veed.io/pricing). The audio track is uploaded and stored in order to transcribe it; the\nvideo is not uploaded. By default OpenEdit does not create a VEED project. The agent guides the one-time\nbrowser login and stores a refreshable token at `veed/.veed-token.json`\n\ninside the runtime.\n\n``` python\nnode --import tsx veed/login.ts\nnode --import tsx veed/go.ts /path/to/video.mp4\n```\n\n**WhisperX** — free, local and offline; nothing leaves your machine. Installed on request, and the first\nrun also downloads a model. CPU-bound on Apple Silicon, with two quality tiers.\n\n```\nbash pipeline/scripts/install-whisperx.sh                  # on request, once\nnode --import tsx prep/transcribe.ts /path/to/video.mp4    # --model medium for the better tier\n```\n\n**Your own service** — produce a Whisper-family JSON however you like (WhisperX, openai-whisper,\nwhisper-timestamped, mlx-whisper, the OpenAI API with `timestamp_granularities=[\"word\"]`\n\n, or whisper.cpp\n`-oj`\n\n) and map it. No credentials pass through OpenEdit.\n\n``` python\nnode --import tsx prep/whisper.ts transcription.json /path/to/video.mp4\n```\n\nPer-word timings are required whichever provider you use: without them the caption reveals drift out of sync with the audio, so a transcript that has none is refused rather than rendered badly.\n\nV1 targets captions. Motion graphics, charts, and brandbook-matched styling render today, but are less exercised than captions and should be expected to have rough edges.\n\nReport defects through GitHub issues.\n\nThe editor is licensed under Apache-2.0. The renderer binaries are distributed under PolyForm Shield\n1.0.0, which permits commercial use of the videos you produce with no payment to VEED. See `LICENSE`\n\nand `NOTICE`\n\nfor the full terms.\n\n**OpenEdit** · powered by [VEED](https://www.veed.io)", "url": "https://wpnews.pro/news/show-hn-openedit-fast-open-source-video-editing-agent", "canonical_source": "https://github.com/veedstudio/open-edit", "published_at": "2026-08-05 16:27:16+00:00", "updated_at": "2026-08-05 16:37:25.202031+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "generative-ai", "developer-tools"], "entities": ["VEED", "OpenEdit", "Claude Code", "Codex", "Gemini", "Fal", "Figma MCP", "Seedance 2.0"], "alternates": {"html": "https://wpnews.pro/news/show-hn-openedit-fast-open-source-video-editing-agent", "markdown": "https://wpnews.pro/news/show-hn-openedit-fast-open-source-video-editing-agent.md", "text": "https://wpnews.pro/news/show-hn-openedit-fast-open-source-video-editing-agent.txt", "jsonld": "https://wpnews.pro/news/show-hn-openedit-fast-open-source-video-editing-agent.jsonld"}}