sipyourdrink-ltd/Bernstein - agent orchestration
Run Bernstein to orchestrate CLI coding agents from GitHub Actions with deterministic, reproducible runs. Supports ad-hoc tasks and automatic CI-fix mode.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| task | Task description for Bernstein, or the literal string "fix-ci" to automatically download failed job logs and attempt a fix. Mutually exclusive with `plan`. | no | "" |
| plan | Path to a Bernstein plan YAML file (e.g. `plans/my-project.yaml`). When set, Bernstein runs the plan with `bernstein run <plan>` instead of a free-text goal. Mutually exclusive with `task`. | no | "" |
| budget | Dollar cap for the run (0 = unlimited). | no | 5.00 |
| cli | Which agent CLI to use (claude, codex, gemini, qwen). | no | claude |
| max-retries | Number of retry attempts in fix-ci mode. | no | 3 |
| python-version | Python version to install. | no | 3.12 |
| post-comment | Post a PR comment with orchestration summary after the run (true/false). | no | true |
Outputs
| name | description |
|---|---|
| tasks-completed | Number of tasks Bernstein completed in this run. |
| total-cost | Total API cost in USD for this run. |
| pr-url | URL of a pull request created or updated by Bernstein (if any). |
| evidence-bundle-path | Workspace-relative path to the evidence bundle (test results, logs, cost report). |