Skip to content

Latest commit

 

History

History
60 lines (46 loc) · 3.18 KB

File metadata and controls

60 lines (46 loc) · 3.18 KB

About OrkasVideoStudio

OrkasVideoStudio turns your coding agent into a video studio. You describe a video in plain language — "make a 60-second vertical explainer on vector databases with a Chinese voiceover and captions" — and the agent reads the material, writes the timeline, and produces the file. Claude Code, Codex, Cursor: any agent that can run a shell or speak MCP can drive it.

Why it exists

Most ways to make video with AI are black boxes: a hosted editor or a one-shot "video agent" hands you a finished file and nothing you can steer. OrkasVideoStudio takes the opposite stance. A video is a readable, diffable, re-renderable plan (plan.json) — change one line and only that piece re-renders. The agent is the brain; this project ships the parts an agent can't improvise:

  • Knowledge — skills that encode what makes a good video and which production line to take, discovered by progressive disclosure.
  • Deterministic capabilities — draft QA / render / edit / transcribe / generate, as auditable wrappers over HyperFrames, ffmpeg, and whisper.cpp.
  • An editable IR — the plan.json timeline that stays yours to inspect and re-render.

What it does — the four production lines

  • Compose — script → designed HTML motion graphics → mp4 (explainers, kinetic type, lower-thirds, data viz, transitions). No paid keys.
  • Edit — cut / join / trim-silence / de-filler / mix / burn-in subtitles / dub / localize real footage you supply, plus highlight selection over long recordings. No paid keys.
  • Generate — talking-head or cinematic AI footage and imagery via your own provider keys (OpenAI, Gemini, Doubao — no managed backend, no lock-in).
  • Auto — when a deliverable needs more than one axis, the agent routes to a single cross-modal plan.json, assembles it segment by segment, and a deterministic delivery guard verifies the finished cut keeps its promise before anything ships.

Who it's for

Reach for OrkasVideoStudio when you want your agent to make the video, keep the result as an auditable file you can edit and re-render, and stay local and open with your own keys — rather than hand-coding every frame in a framework or handing the job to a hosted SaaS editor.

Where it comes from

OrkasVideoStudio began as the built-in video agent inside Orkas, the AI-team desktop app, where it was validated end-to-end — then extracted into this agent-agnostic, MIT-licensed toolkit. Inside Orkas it ships built in; everywhere else you install it into your own coding agent and get the same capabilities via the ovs CLI and MCP server.

Links: orkas.ai · github.com/Orkas-AI · repository

Learn more

  • README — install, use cases, the full command surface, and how it compares.
  • PLAN.md — the technical plan and roadmap.
  • AGENTS.md — how a coding agent picks up the skills and capabilities.

MIT licensed — see LICENSE.