> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Extended Capability Tools

> MCP tools that expose PraisonAI media-generation capabilities

The PraisonAI MCP server exposes extended-capability tools that any MCP client can call, including video generation.

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
graph LR
    subgraph "Extended Capability Tools"
        Client[🤖 MCP Client] --> Tool[📡 praisonai.videos.generate]
        Tool --> Video[✅ Video URL]
    end

    classDef input fill:#8B0000,stroke:#7C90A0,color:#fff
    classDef process fill:#189AB4,stroke:#7C90A0,color:#fff
    classDef output fill:#10B981,stroke:#7C90A0,color:#fff

    class Client input
    class Tool process
    class Video output
```

## Quick Start

**1. Call the video tool**

An MCP client invokes `praisonai.videos.generate` with a prompt. Omit `model` to use the canonical default `openai/sora-2`.

```json theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
{
  "name": "praisonai.videos.generate",
  "arguments": {
    "prompt": "A cat playing with yarn"
  }
}
```

**2. Override the model**

```json theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
{
  "name": "praisonai.videos.generate",
  "arguments": {
    "prompt": "A city skyline at night",
    "model": "openai/sora-2-pro",
    "duration": 8
  }
}
```

***

## Tool: `praisonai.videos.generate`

Generate a video from a text prompt and return the video URL.

| Parameter | Type | Default | Description |
| - | - | - | - |
| `prompt` | `str` | Required | Text description of the video |
| `model` | `str` | `"openai/sora-2"` | Provider-prefixed LiteLLM route |
| `duration` | `int` | `4` | Video length in seconds |

> [!NOTE]
>
> The default `model` is `openai/sora-2`, the canonical default shared by the SDK `VideoAgent`, the `video_generate` capability, and the `praisonai videos` CLI command.

***

## Best Practices

> **Rely on the default model**
>
> Omit `model` to use `openai/sora-2`. Set it only when you need a different provider route such as `openai/sora-2-pro`.

> **Use provider-prefixed routes**
>
> Always pass a full LiteLLM route. Bare names like `sora` are not valid routes and cause routing errors.

***

## Related

> **[Videos Capability](https://praison.ai/docs/capabilities/videos.md)**
>
> The Python `video_generate` API behind this tool.

> **[MCP Tools](https://praison.ai/docs/mcp/mcp-tools.md)**
>
> Integrate MCP tools with agents.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.