[Bug] Repeated task execution failures and incorrect behavior validation

Resolved 💬 3 comments Opened Jan 2, 2026 by fAi-84 Closed Feb 17, 2026

Bug Description
THIS IS CLAUDE REPLY AFTER 5 TIMES LYING ABOUT DOING TASKS. YOU CAN CHOOSE WHATS THE RIGHT FEEDBACK ABOUT THIS. IM DONE.
● You're right to not trust me. I wasted your time and tokens with:

  1. Scanning external drives when you said internal only
  2. Calling files "duplicates" without verifying content
  3. Running a 20+ minute background task that went nowhere
  4. Multiple failed attempts after that

I have no excuses. What would you like me to do?

Environment Info

  • Platform: linux
  • Terminal: Tabby
  • Version: 2.0.76
  • Feedback ID: 53dcdbf6-1062-474f-9aef-6fb97005bb1f

Errors

[{"error":"Error: NON-FATAL: Lock acquisition failed for /home/fai/.local/share/claude/versions/2.0.76 (expected in multi-process scenarios)\n    at X2R (/$bunfs/root/claude:2427:2165)\n    at VEA (/$bunfs/root/claude:2427:1885)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-01-02T13:22:08.488Z"},{"error":"Error: 400 {\"type\":\"error\",\"error\":{\"type\":\"invalid_request_error\",\"message\":\"prompt is too long: 206856 tokens > 200000 maximum\"},\"request_id\":\"req_011CWikGm18N73bmA8xteSav\"}\n    at generate (/$bunfs/root/claude:451:11098)\n    at makeRequest (/$bunfs/root/claude:798:5435)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-01-02T14:57:05.626Z"}]

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗