Nemo Video

Integration

OpenAI Codex integration

Integrate NemoVideo with OpenAI Codex for automated video generation in your CI/CD pipelines, scripts, and autonomous coding workflows.

Setup

Add NemoVideo as a tool in your Codex environment by including the configuration in your project: Get your API key here.

{
  "tools": [
    {
      "name": "nemovideo",
      "type": "skill",
      "source": "nemovideo/nemovideo-tools",
      "config": {
        "api_key_env": "NEMOVIDEO_API_KEY"
      }
    }
  ]
}

Note · Make sure to set the NEMOVIDEO_API_KEY environment variable in your Codex workspace settings.

Usage

Once configured, Codex can generate videos as part of its task execution. Here are common use cases:

Automated content generation

Read the CHANGELOG.md and generate a 15-second product update video highlighting the top 3 new features. Use a professional style with our brand colors.

CI/CD integration

Generate demo videos automatically when PRs are merged or releases are tagged:

name: Generate Release Video
on:
  release:
    types: [published]

jobs:
  generate-video:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Generate release demo
        env:
          NEMOVIDEO_API_KEY: ${{ secrets.NEMOVIDEO_API_KEY }}
        run: |
          codex run "Generate a professional product demo
          video showcasing the new features. 16:9, 15s."

Batch processing

For each product image in the /assets folder, generate a short promotional video. Use 9:16 aspect ratio for social media. Save outputs to /videos.

Best Practices

01

Use explicit prompts

Include duration, aspect ratio, style, and quality parameters explicitly for deterministic results.

02

Handle async tasks

Video generation is asynchronous. Codex will poll for results automatically, but set appropriate timeouts.

03

Manage API limits

Be mindful of rate limits in batch operations. The tool handles retries with exponential backoff.

04

Version your prompts

Store video generation prompts alongside your code for reproducibility and iteration.