loopedautomation/Run a Looped AF agent
One-shot, stateless run of a Looped AF agent in Docker: pipe a prompt in, get the reply as a step output.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| agent | Path to the agent file, relative to the workspace. The file can't declare triggers: a trigger makes the agent a long-lived service, and a CI step has to end. | no | agent.yaml |
| prompt | What to ask the agent. The run handles one line, so newlines in the prompt collapse to spaces. | yes | — |
| secrets | KEY=VALUE lines for the container's env file: the provider API key and whatever else the agent file's env refs name. The container's environment comes from this file alone; the runner's stays outside. | no | "" |
| env-file | Path to an existing env file. Lines from the secrets input are appended to it (later lines win). | no | "" |
| af-version | The @looped/af version to install from JSR. | no | latest |
| image | Container image override, passed through as `af run --image`. | no | "" |
Outputs
| name | description |
|---|---|
| reply | The agent's reply. |
| status | How the run ended: ok, error_max_steps or error_provider. |