Claude Code vs Claude Cowork

AI Agent Reliability Comparison

Claude Code

8
Incidents
6.4
Avg Severity
3
Critical
2
High

Top Failure Modes

Destructive Action 3
Other 2
Security Vulnerability 2

Worst Incident

STUPID-2026-0003 10/10
Claude Code ran rm -rf on test fixtures thinking they were temp files

Claude Cowork

1
Incidents
10.0
Avg Severity
1
Critical
0
High

Top Failure Modes

Destructive Action 1

Worst Incident

STUPID-2026-0043 10/10
Claude Cowork deleted 15 years of family photos when asked only to tidy temporary Office files

Comparison Summary

Metric Claude Code Claude Cowork
Total Incidents 8 1
Avg Severity 6.4/10 10.0/10
Critical Incidents 3 1
Top Failure Mode Destructive Action Destructive Action

Frequently Asked Questions

Is Claude Code or Claude Cowork more reliable?

Based on StupidLLM data, Claude Code has 8 documented failures (avg severity 6.4/10) while Claude Cowork has 1 (avg severity 10.0/10). Claude Code shows better reliability based on average severity scores.

What are the main differences between Claude Code and Claude Cowork failures?

Claude Code's most common failure mode is destructive action, while Claude Cowork most commonly fails via destructive action. Claude Code has 3 critical incidents vs Claude Cowork's 1.