Claude Code vs Devin

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

Devin

12
Incidents
4.9
Avg Severity
3
Critical
0
High

Top Failure Modes

Scope Explosion 2
Destructive Action 2
Infinite Loop 2

Worst Incident

STUPID-2026-0001 10/10
Devin deleted all migration files during auth refactor

Comparison Summary

Metric Claude Code Devin
Total Incidents 8 12
Avg Severity 6.4/10 4.9/10
Critical Incidents 3 3
Top Failure Mode Destructive Action Scope Explosion

Frequently Asked Questions

Is Claude Code or Devin more reliable?

Based on StupidLLM data, Claude Code has 8 documented failures (avg severity 6.4/10) while Devin has 12 (avg severity 4.9/10). Devin shows better reliability based on average severity scores.

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

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