Setup

Get Started

Spin up OpenVideoKit locally in under 5 minutes. Python 3.13+ required.

1.Install & run

Create a venv, install, and boot. The bundled wheel ships the SPA, all 25 slide templates, edge-tts, MP4 export, the LangGraph AI copilot, and the MCP endpoint — no separate frontend build needed.

bash

python3 -m venv .venv
source .venv/bin/activate           # Windows: .venv\Scripts\activate
pip install openvideokit
ovk serve                           # → http://localhost:8000

Open http://localhost:8000— you're in the editor.

2.Enable the AI copilot(optional)

On first ovk serve, OpenVideoKit writes a documented config file to~/.openvideokit/config.toml(chmod 600 — secrets safe). Open it and add your OpenAI-compatible API key:

bash

ovk config edit

Edit the [ai]section. Works with OpenAI, OpenRouter, Ollama, vLLM, LM Studio:

bash

[ai]
openai_base_url = "https://api.openai.com/v1"
openai_api_key  = "sk-proj-..."
model           = "gpt-5.4-nano"

Stop the server (Ctrl + C) and re-run ovk serve. Verify the connection:

bash

ovk llm test

3.Connect Claude Code / Cursor / Goose(optional)

The MCP endpoint is live athttp://localhost:8000/mcp/by default. Point any MCP-compatible client at it:

json

{
  "mcpServers": {
    "openvideokit": {
      "url": "http://localhost:8000/mcp/"
    }
  }
}

Then ask your agent: "Create a 3-slide product video and set the title on each slide."Mutations go through the same 28-tool API the in-app copilot uses — every change persists to disk immediately.

Next steps

Browse the 25-template gallery — live video previews, field defaults, copy-ready HTML
Full config reference — priority chain (env > .env > config.toml > defaults), CLI commands
AI copilot deep dive — LangGraph agent, EditOp flow, prompt caching
MCP design doc — execute-on-call semantics, all 28 tools

For contributors

If you want to hack on the source:

bash

git clone https://github.com/quanhua92/openvideokit
cd openvideokit
uv sync --extra dev
cd ovk-web && pnpm install && cd ..

# Terminal 1 — Python API + live reload
uv run ovk serve --reload

# Terminal 2 — Vite dev (proxies /api → :8000)
cd ovk-web && pnpm dev    # → http://localhost:3000

Edit .env(from .env.example) to set OPENAI_API_KEYfor AI development.