[BUG] Context limit reached repeatedly during multi-agent tasks prevents completion

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?

Bug Description

● Task Output afbbe80
  ⎿  Read output (ctrl+o to expand)
  ⎿  Context limit reached · /compact or /clear to continue

✻ Sautéed for 11m 9s

● Agent "Audit parsing" completed
  ⎿  Context limit reached · /compact or /clear to continue

● Agent "Audit code" completed
  ⎿  Context limit reached · /compact or /clear to continue

● Agent "Audit internals" completed
  ⎿  Context limit reached · /compact or /clear to continue

● Agent "Audit xyz" completed
  ⎿  Context limit reached · /compact or /clear to continue

● Agent "Audit abc" completed
  ⎿  Context limit reached · /compact or /clear to continue

What Should Happen?

Claude should have completed the audit as requested.

Error Messages/Logs

Steps to Reproduce

I can't provide exact STR.

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

pre Opus-4.6

Claude Code Version

2.1.34 (Claude Code)

Platform

Anthropic API

Operating System

Other Linux

Terminal/Shell

Xterm

Additional Information

_No response_