---
title: "FFmpeg Agent"
description: "Local-first visual environment for building and running AI workflows. Build agents visually, deploy anywhere, privacy by design."
canonical: https://docs.nodetool.ai/nodes/nodetool/agents/ffmpegagent
markdown: https://docs.nodetool.ai/nodes/nodetool/agents/ffmpegagent.md
product: NodeTool
source: https://github.com/nodetool-ai/nodetool/blob/main/docs/nodes/nodetool/agents/ffmpegagent.md
---

# FFmpeg Agent

**Type:** `nodetool.agents.FfmpegAgent`

**Namespace:** `nodetool.agents`

## Description

Edit, convert, and package audio and video with FFmpeg from a prompt.
    agent, ffmpeg, media, video, audio, convert

## Properties

| Property | Type | Description | Default |
|----------|------|-------------|---------|
| model | `language_model` | Model used for media prompts. | `{"type":"language_model","provider":"empty","id...` |
| audio | `list[audio]` | Optional audio inputs for media reasoning tasks. Accepts a list, or a single Audio (auto-wrapped). | `[]` |
| video | `list[video]` | Optional video inputs for media reasoning tasks. Accepts a list, or a single Video (auto-wrapped). | `[]` |
| prompt | `str` | Prompt for media task execution. | `` |
| timeout_seconds | `int` | Maximum runtime for agent execution. | `180` |
| max_output_chars | `int` | Maximum serialized output chars before truncation. | `200000` |

## Outputs

| Output | Type | Description |
|--------|------|-------------|
| video | `video` |  |
| audio | `audio` |  |
| text | `str` |  |

## Related Nodes

Browse other nodes in the [nodetool.agents](./) namespace.