![\"CursorCam\"](\"./assets/logo-wordmark-dark.svg\"){kind=logo}
\n \n ![\"npm\"](\"https://shieldcn.dev/npm/v/cursorcam.svg?variant=secondary&size=xs&theme=zinc\")
\n ![\"GitHub](\"https://shieldcn.dev/github/stars/Avijit07x/cursorcam.svg?variant=secondary&size=xs&theme=zinc\")
\n ![\"Contributors\"](\"https://shieldcn.dev/github/contributors/Avijit07x/cursorcam.svg?variant=secondary&size=xs&theme=zinc\")
\n ![\"Last](\"https://shieldcn.dev/github/last-commit/Avijit07x/cursorcam.svg?variant=secondary&size=xs&theme=zinc\")
\n ![\"License](\"https://shieldcn.dev/github/license/Avijit07x/cursorcam.svg?variant=secondary&size=xs&theme=zinc\")
\n
![\"A](\"./assets/demo.webp\")
\n\nMade with CursorCam from 13 steps. Watch the full-quality MP4 (1080p, 60 fps).
\n\n---\n\n## Setup\n\n**You need** Node.js 22.13+ and Chrome, Edge, Chromium or Brave 120+. CursorCam uses the browser you already have and downloads nothing.\n\n**Install it in your agent:**\n\n| Agent | Install |\n| --- | --- |\n| Claude Code | `/plugin marketplace add Avijit07x/cursorcam`, then `/plugin install cursorcam@cursorcam` and `/reload-plugins` |\n| Codex | `codex plugin marketplace add Avijit07x/cursorcam`, then `codex plugin add cursorcam@cursorcam` |\n| VS Code and GitHub Copilot | Add `Avijit07x/cursorcam` to `chat.plugins.marketplaces`, then install it from `@agentPlugins` |\n| Gemini CLI | `gemini extensions install https://github.com/Avijit07x/cursorcam` |\n| Cursor | [One-click MCP install](./docs/agents.md#cursor) |\n| Claude Desktop | Open [`cursorcam.mcpb`](https://github.com/Avijit07x/cursorcam/releases/latest/download/cursorcam.mcpb) |\n| Any other app | The MCP server, `npx -y cursorcam mcp`. See the [MCP guide](./docs/mcp.md) |\n\n[Your AI agent](./docs/agents.md) has each one step by step.\n\n**Then ask for a video:**\n\n```\nMake a 20 second demo of http://localhost:3000: sign up and create a project.\n```\n\nYou get `video.mp4` (1080p, 60 fps) and a `poster.jpg` thumbnail in a run folder inside `cursorcam-output/`. Add that folder to `.gitignore`.\n\n## How it works\n\nYour agent explores your app, writes the steps and records them in a hidden browser. Then it checks stills of the video and fixes anything that looks wrong.\n\nIf it needs a login or a 2FA code, it asks you. What you type is used for that run only and never saved.\n\n## Use the CLI\n\nThe MCP server and the plugins run the `cursorcam` CLI. You can run it yourself too:\n\n```bash\nnpx cursorcam doctor\nnpx cursorcam run steps.json\n```\n\n```json\n{\n \"url\": \"http://localhost:3000\",\n \"steps\": [\n { \"click\": \"role=button[name='Get started']\" },\n { \"type\": \"alex@example.com\", \"into\": \"label=Email\" },\n { \"press\": \"Enter\" },\n { \"waitFor\": \"text=Dashboard\" }\n ]\n}\n```\n\n## Docs\n\n| Guide | Covers |\n| --- | --- |\n| [Getting started](./docs/getting-started.md) | Set up CursorCam and make your first video |\n| [Your AI agent](./docs/agents.md) | Install it in Claude Code, Codex, VS Code, Gemini CLI, Cursor or Claude Desktop |\n| [MCP server](./docs/mcp.md) | Set it up by hand in any MCP app, and every tool it offers |\n| [Better videos](./docs/better-videos.md) | How to ask your AI agent, with example requests and tips |\n| [Steps file](./docs/steps.md) | Every action, target and option |\n| [Look and size](./docs/style.md) | Presets for X, LinkedIn, YouTube and Discord, formats, backgrounds and zoom |\n| [Logins and secrets](./docs/logins.md) | Passwords, 2FA codes and saved logins |\n| [CLI](./docs/cli.md) | Every command, the run folder and exit codes |\n| [Troubleshooting](./docs/troubleshooting.md) | Common errors and how to fix them |\n| [Examples](./docs/examples/) | Steps and style files to copy and edit |\n\n## Contributing\n\nBug reports and pull requests are welcome. See [CONTRIBUTING.md](./CONTRIBUTING.md). To report a security problem, see [SECURITY.md](./SECURITY.md).\n\n## License\n\n[Apache-2.0](./LICENSE)\n","readmeFilename":"README.md"}