list-files must include at least the first-level directory contents

[qdaxb]

Related GitHub Issue

Closes: #5301

Description

When asking the model to “summarize the current project” in projects with a large number of files, it may make inferences based on the content returned by the Current Workspace Directory or the List Files tool.

However, the current List-files tool uses a depth-first search strategy, which can result in returning detailed structures for only part of the project directories, while the model has no awareness of the other directories at all. In some weaker models, instead of requesting the agent to fetch all subfolders, they directly infer based on the partial information available, which can lead to completely incorrect results.

Changes:

• If the number of files retrieved by list-files reaches the limit, perform an additional retrieval of all directories and files under the top-level folders separately, then merge the results before proceeding with further processing.
• formatAndCombineResults uses a breadth-first traversal approach.
• (Note: listFilesWithRipgrep still uses depth-first search, which may lead to incomplete results in some cases, but at least the first-level directory structure will not be missing.)

Test Procedure

When viewing the prompt in human-relay mode, the Current Workspace Directory will include at least all files in the first-level directories.

Pre-Submission Checklist

• Issue Linked: This PR is linked to an approved GitHub Issue (see "Related GitHub Issue" above).
• Scope: My changes are focused on the linked issue (one major feature/fix per PR).
• Self-Review: I have performed a thorough self-review of my code.
• Testing: New and/or updated tests have been added to cover my changes (if applicable).
• Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
• Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Screenshots / Videos

Documentation Updates

Additional Notes

Get in Touch

---

Important

Enhance listFiles to include first-level directory contents when file limit is reached, ensuring complete directory awareness.

• Behavior:
  • listFiles in list-files.ts now performs an additional non-recursive retrieval of top-level directories if the file limit is reached in recursive mode.
  • Ensures at least first-level directory contents are included in results.
• Functions:
  • formatAndCombineResults updated to sort paths by depth, ensuring directories are listed before files.
• Misc:
  • listFilesWithRipgrep still uses depth-first search, but first-level directory structure is guaranteed.

^{This description was created by for e64dd2a743d22008ac788e5e4806ed3a738759b6. You can customize this summary. It will automatically update as commits are pushed.}

[daniel-lxs]

Thank you @qdaxb for your contribution!

LGTM