Issue Station
Sign In
Tools hereToolbox

claude-code - 💡(How to fix) Fix TypeError: null is not an object (evaluating 'T.effortLevel') - sessions interrupted on every prompt [1 comments, 1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#51896•Fetched 2026-04-23 07:42:00
View on GitHub
Comments
1
Participants
1
Timeline
7
Reactions
0
Author
ewasarenkaewasarenka
Participants
ewasarenkaewasarenka
Timeline (top)
labeled ×4closed ×1commented ×1cross-referenced ×1

Error Message

[ERROR] TypeError: TypeError: null is not an object (evaluating 'T.effortLevel') at JEA (/$bunfs/root/claude:1780:3906) at aK (/$bunfs/root/claude:2142:4153) at processTicksAndRejections (native:7:39)

[ERROR] Error: Error: 1: 1 at H57 (/$bunfs/root/claude:1025:4019) at async B57 (/$bunfs/root/claude:1025:1623)

[ERROR] Error: NON-FATAL: Lock acquisition failed for ~/.local/share/claude/versions/2.0.76 (expected in multi-process scenarios)

Code Example

⎿  Interrupted · What should Claude do instead?

---

[ERROR] TypeError: TypeError: null is not an object (evaluating 'T.effortLevel')
    at JEA (/$bunfs/root/claude:1780:3906)
    at aK (/$bunfs/root/claude:2142:4153)
    at processTicksAndRejections (native:7:39)

[ERROR] Error: Error: 1: 1 
    at H57 (/$bunfs/root/claude:1025:4019)
    at async B57 (/$bunfs/root/claude:1025:1623)

[ERROR] Error: NON-FATAL: Lock acquisition failed for ~/.local/share/claude/versions/2.0.76 (expected in multi-process scenarios)
RAW_BUFFERClick to expand / collapse

Bug Description

New Claude Code sessions get interrupted on every prompt with the message:

⎿  Interrupted · What should Claude do instead?

This happens immediately after typing any prompt (even just "hi"). Existing sessions continue to work, but no new sessions can be started.

Debug Log Errors

[ERROR] TypeError: TypeError: null is not an object (evaluating 'T.effortLevel')
    at JEA (/$bunfs/root/claude:1780:3906)
    at aK (/$bunfs/root/claude:2142:4153)
    at processTicksAndRejections (native:7:39)

[ERROR] Error: Error: 1: 1 
    at H57 (/$bunfs/root/claude:1025:4019)
    at async B57 (/$bunfs/root/claude:1025:1623)

[ERROR] Error: NON-FATAL: Lock acquisition failed for ~/.local/share/claude/versions/2.0.76 (expected in multi-process scenarios)

The effortLevel error appears multiple times during session initialization.

Troubleshooting Attempted

  • Removed ~/.claude/plugins/marketplaces/claude-plugins-official (was corrupted)
  • Backed up and removed project session data (~/.claude/projects/...)
  • Backed up and removed ~/.claude.json
  • Killed stale Claude processes

None of these resolved the issue.

Environment

  • Claude Code version: 2.0.76
  • OS: macOS (Darwin 24.6.0)
  • Plan: Claude Team
  • Model: Opus 4.5

Steps to Reproduce

  1. Have an existing Claude session running
  2. Open a new terminal
  3. Run claude or claude --debug
  4. Type any prompt
  5. Observe immediate interruption

Expected Behavior

New sessions should accept prompts and respond normally.

Actual Behavior

Every prompt is immediately interrupted with "What should Claude do instead?"

extent analysis

TL;DR

The issue is likely related to a null object error when evaluating 'effortLevel' during session initialization, and a potential workaround is to investigate and resolve the effortLevel null value issue.

Guidance

  • Investigate the effortLevel variable and its initialization process to determine why it is null, as this is the primary error message.
  • Review the code at /$bunfs/root/claude:1780:3906 to understand how effortLevel is being used and why it might be null.
  • Consider the Lock acquisition failed error as a potential contributor to the issue, especially in multi-process scenarios, and explore ways to resolve this lock acquisition failure.
  • Given the removal of various files and killing of stale processes did not resolve the issue, it may be necessary to delve deeper into the application's state and configuration to identify the root cause.

Example

No specific code example can be provided without more context on the effortLevel variable and its usage within the Claude Code application.

Notes

The provided debug log errors and troubleshooting attempts suggest a complex issue that may require a deeper dive into the application's code and configuration. The fact that existing sessions continue to work but new ones cannot be started implies a potential issue with session initialization or resource allocation.

Recommendation

Apply workaround: Given the information provided, it seems that applying a workaround or fix related to the effortLevel null object error and the lock acquisition failure might be necessary. Upgrading to a potentially fixed version is not explicitly implied as an option from the provided information.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#installation#tensor shape#autograd error#model save/load

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

claude-code - 💡(How to fix) Fix TypeError: null is not an object (evaluating 'T.effortLevel') - sessions interrupted on every prompt [1 comments, 1 participants]