icoretech/Codex Action
Run OpenAI Codex CLI non-interactively in GitHub Actions workflows
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| prompt | Instructions for Codex (e.g., "Summarize these changes for operators") | yes | — |
| input_text | Data to process (e.g., changelog content). Appended after the prompt with a separator. | no | "" |
| openai_api_key | OpenAI API key. Mutually exclusive with codex_config. | no | "" |
| codex_config | Base64-encoded auth.json from a prior device-auth session. Mutually exclusive with openai_api_key. | no | "" |
| codex_config_toml | Base64-encoded config.toml with Codex preferences (model, personality, etc.). Optional, works with either auth method. | no | "" |
| image_version | codex-docker image version tag | no | 0.145.0 |
| model | Model override (passed to codex exec --model) | no | "" |
| reasoning_effort | Reasoning effort level (minimal, low, medium, high, xhigh). Passed via model_reasoning_effort config override. | no | "" |
| network_access | Allow Codex to make network requests (curl, wget, etc.) during execution. When false (default), a prompt-level policy instructs the model not to use networking tools. | no | false |
| sandbox | Sandbox mode for Codex execution. Options: full-auto (default, uses bubblewrap), danger-full-access (disables sandbox — recommended for CI/Docker where the container is already isolated). | no | full-auto |
| quiet | Suppress verbose Codex output (tool calls, grep results, file reads) from workflow logs. Prevents source code leakage in logs. Set to false for debugging. | no | true |
| timeout | Max seconds for codex execution (default: 300) | no | 300 |
Outputs
| name | description |
|---|---|
| result | Text output from Codex |