Skip to content

feat: Add agent model display and configuration improvements#470

Open
szmania wants to merge 7 commits intodwash96:v0.98.0from
szmania:cli-2-agent-model
Open

feat: Add agent model display and configuration improvements#470
szmania wants to merge 7 commits intodwash96:v0.98.0from
szmania:cli-2-agent-model

Conversation

@szmania
Copy link
Copy Markdown

@szmania szmania commented Mar 25, 2026

Summary

This PR adds comprehensive agent model display and configuration improvements to enhance the agent mode experience in cecli.

Key Changes

Agent Model Display

  • Footer Updates: The footer now displays the active agent model name when in agent mode
  • Spinner Text: Updated spinner text to show the active agent model for better user feedback
  • TUI Integration: Enhanced TUI components to properly display agent model information

Agent Model Configuration

  • Fallback Handling: AgentCoder now uses main_model as a fallback when agent_model is not explicitly set
  • Async Generator: Fixed AgentCoder.send to be an async generator for proper async handling
  • Model Name Resolution: Added get_active_model_name() method to correctly identify the active model

Code Quality Improvements

  • Import Fixes: Resolved import issues in test files for conversation system
  • Language Normalization: Fixed C# language identifier normalization in repomap
  • Code Cleanup: Removed deprecated workspace_paths configuration and related code

Testing

  • Test Fixes: Updated test_conversation_system.py to use correct import paths
  • C# Support: Fixed C# symbol extraction in repomap tests
  • All Tests: Verified all tests pass in the basic test suite

Technical Details

  • Files Modified:
    • cecli/coders/base_coder.py: Added agent model display logic
    • cecli/main.py: Removed deprecated workspace_paths configuration
    • cecli/repomap.py: Fixed C# language identifier handling
    • cecli/args.py: Cleaned up workspace_paths argument handling
    • cecli/args_formatter.py: Removed workspace_paths documentation
    • tests/test_conversation_system.py: Fixed import statements

Testing

All changes have been tested and verified to work correctly. The agent model display now provides clear feedback about which model is active during agent mode operations.

Breaking Changes

None - all changes are backward compatible and focus on improving the user experience in agent mode.

szmania added 7 commits March 24, 2026 21:31
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
Co-authored-by: cecli (openai/openai_gemini_cli/gemini-2.5-pro)
@dwash96 dwash96 changed the base branch from main to v0.98.0 March 28, 2026 04:36
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