$ docker-agent eval fixtures/agents/files.yaml fixtures/evals --judge-model dmr/ai/qwen3:4b -c 1 --output work/eval-results
Evaluation run: <run>
Loading evaluation sessions...
Validating judge model...
Pre-building 1 container image(s)...
Running 2 evaluations with concurrency 1

✗ Count the lines of README.md ($0.000000)
  ✗ running docker agent in container: container failed: exit status 1 (stderr: Error: failed to get models: docker --config=/root/.docker --context=default model status --json: : exec: "docker": executable file not found in $PATH)
✗ List the files in the working directory ($0.000000)
  ✗ running docker agent in container: container failed: exit status 1 (stderr: Error: failed to get models: docker --config=/root/.docker --context=default model status --json: : exec: "docker": executable file not found in $PATH)
[██████████] 100% (2/2) ✓0 ✗2

❌         Errors: 2/2 evaluations failed

Total Cost: $0.000000
Total Time: <dur>

Sessions DB: $CAPTURE/work/eval-results/<run>.db
Sessions JSON: $CAPTURE/work/eval-results/<run>.json
Log: $CAPTURE/work/eval-results/<run>.log
[exit 0]