Skip to content

fix: use correct --model flag for pi coding agent#253

Open
Spacewalker215 wants to merge 1 commit intostephengpope:mainfrom
Spacewalker215:patch-1
Open

fix: use correct --model flag for pi coding agent#253
Spacewalker215 wants to merge 1 commit intostephengpope:mainfrom
Spacewalker215:patch-1

Conversation

@Spacewalker215
Copy link
Copy Markdown

Description

This PR fixes a bug in the Pi coding agent's startup script that causes it to crash immediately when an LLM_MODEL override is provided.

What changed:
Changed the -m flag to --model in docker/coding-agent/scripts/agents/pi-coding-agent/run.sh.

Why:
The pi CLI does not recognize the -m shorthand. When a headless task or cluster worker runs with a specific model configured, the script previously passed (-m "$LLM_MODEL"), which resulted in a fatal Error: Unknown option: -m crash. The interactive script (interactive.sh) already correctly uses --model, so this brings run.sh into alignment.

Type of Change

  • Bug fix (fix:)
  • New feature (feat:)
  • Documentation update (docs:)
  • Refactoring (refactor:)

Testing

Verified that pi successfully accepts the --model flag without throwing an unknown option error, allowing the container to proceed with execution.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant