Skip to content

Conversation

@code-yeongyu
Copy link
Owner

Background

OmO agent was being too eager - it would immediately create TODO lists and start implementing whenever it received any request, even when users were just exploring, asking questions, or hadn't explicitly asked for implementation. This violated the principle of letting users control when actual work begins.

Changes

Added explicit behavioral constraints to the OmO agent prompt in src/agents/omo.ts:

  • Core Competencies section: Added instruction to never start implementing unless the user explicitly requests it, with a reminder that TODO creation is tracked by hooks
  • Task Management workflow: Clarified that TODOs should only be created for implementation tasks when the user actually wants implementation

The constraint is emphasized at two strategic points in the prompt to ensure the agent internalizes this behavior.

Review Notes

  • The branch name omo-dontwork reflects the intent: OmO should "don't work" (not proactively work) unless asked
  • This is a prompt-level behavioral change, no code logic was modified
  • The change aligns with better UX where users maintain control over when AI takes action

🤖 GENERATED WITH ASSISTANCE OF OhMyOpenCode

…ermission check (#89)

- Increase delay from 200ms to 5000ms to prevent firing too quickly before users can respond
- Add write permission check to skip continuation when previous agent lacks write/edit permissions
- Fixes destructive behavior where hook was overriding user wait commands

Resolves #89

🤖 Generated with assistance of OhMyOpenCode (https://github.com/code-yeongyu/oh-my-opencode)
@code-yeongyu code-yeongyu merged commit df0a9e6 into master Dec 18, 2025
1 check passed
@code-yeongyu code-yeongyu deleted the omo-dontwork branch December 18, 2025 16:45
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.

2 participants