Codex CLI Integration with Veo MCP
📘 Complete Guide: Using all 11 AceData Cloud MCPs in Codex CLI →
Google DeepMind's video model supports Veo 2 / Veo 3 / Veo 3.1, with optional 1080P + native audio. After integrating MCP with Codex CLI, you can directly invoke it in conversations without switching windows or copy-pasting.
¶ Obtaining API Token
Register an account at AceData Cloud Platform, and copy the API Token from the homepage — one token can be used for all 11 MCP Servers. Detailed steps are available in the Complete Guide.
¶ Configuration
Edit the configuration file: ~/.codex/config.toml
[mcp_servers.veo]
url = "https://veo.mcp.acedata.cloud/mcp"
type = "http"
http_headers = { "Authorization" = "Bearer yourToken" }
After saving, the tool will be automatically loaded on the next codex session startup.
¶ Usage Examples
Generate an 8-second video with Veo 3: morning in the mountains, sea of clouds rolling, with natural ambient sound
Use this image (link xxx) as the first frame to generate an 8-second camera movement video
Quickly generate a product demo video with Veo 3.1 fast, 1080P
¶ Tools
| Tool | Description |
|---|---|
| veo_text_to_video | Veo 2 / 3 / 3.1 text-to-video |
| veo_image_to_video | image-to-video |