[BUG] Stale Diagnostics?

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

Often times I will get messages like this:

  Wave 1 Complete

  04.1-01: Refactor get_job() to reuse _build_job_response() + add tests
  get_job() now calls _build_job_response() and returns JobDetailResponse (16 fields instead of 10). 6 new tests in
 pTestGetJobDetailEndpoint. 1 auto-fix for MagicMock name parameter gotcha.

  However, I see diagnostics flagging an unused JobDetailResponse import in jobs.py and some unused parameters in tests. Let me check and fix the import issue before proceeding to verification.

● Read(src/api/routes/jobs.py)
  ⎿  Read 40 lines

● Search(pattern: "JobDetailResponse", path: "src/api/routes/jobs.py", output_mode: "content")
  ⎿  Found 5 lines (ctrl+o to expand)

● The import IS used on lines 404, 412, 426, 449. The diagnostic is a false positive — pyright might be analyzing the file before the executor's changes were saved. Let me verify the current state by running the checks.

I am using VSCode and have the pyright LSP plugin enabled. VSCode is set to autosave with a short delay. Is this an issue with the LSP server or something else?

Has anyone experienced anything like this? It seems to happen often.

What Should Happen?

Diagnostics should accurately reflect the file as it is and not return stale false positives

Error Messages/Logs

Steps to Reproduce

1. Install pyright lsp plugin
2. Have CC update files that add new imports
3. CC gets stale diagnostic messages

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.20

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

WSL (Windows Subsystem for Linux)

Additional Information

_No response_