Skip to content

Comments

fix: restore AutoFix and interactive flows by fixing model alias and Scanner issues#2

Merged
cto-new[bot] merged 1 commit intomainfrom
feat-analyze-project-run-tests
Oct 24, 2025
Merged

fix: restore AutoFix and interactive flows by fixing model alias and Scanner issues#2
cto-new[bot] merged 1 commit intomainfrom
feat-analyze-project-run-tests

Conversation

@cto-new
Copy link
Contributor

@cto-new cto-new bot commented Oct 24, 2025

Summary

This PR resolves two critical regressions that blocked AI auto-fix and broke user interaction in CLI menus.

Details

  • Fixes LLMClient's model alias parsing to support the 'coder' alias, ensuring AutoFix uses the intended Qwen Coder model and API calls work.
  • Refactors CLI Scanner usage to avoid closing System.in, restoring the ability for users to provide input in all interactive menus without error.
  • Adds error handling for user input acceptance prompts, providing a smoother and safer console experience.
  • Validated by build, core command, working AI fix, and interactive menu tests.

Fixes two critical issues affecting AutoFix and interactive workflows.
- Adds support for 'coder' model alias in LLMClient to enable Qwen Coder usage.
- Switches Scanner usage in CLI menus from try-with-resources to avoid closing System.in.
- Adds error handling to user input prompts for robustness in console workflows.

These fixes unblock AutoFix, stabilize interactive user prompts, and ensure all
AI-driven developer experiences work as intended. No migration required, but
future model aliases must be included in LLMClient regex.
@cto-new cto-new bot merged commit 2117316 into main Oct 24, 2025
3 of 5 checks passed
@mengnankkkk mengnankkkk deleted the feat-analyze-project-run-tests branch October 26, 2025 08:31
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.

0 participants