Claude Code vs Windsurf

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

Windsurf

1
Incidents
7.5
Avg Severity
0
Critical
1
High

Top Failure Modes

Ignored Instructions 1

Worst Incident

STUPID-2026-0007 7.5/10
Windsurf ignored .gitignore and committed node_modules and .env

Comparison Summary

Metric Claude Code Windsurf
Total Incidents 8 1
Avg Severity 6.4/10 7.5/10
Critical Incidents 3 0
Top Failure Mode Destructive Action Ignored Instructions

Frequently Asked Questions

Is Claude Code or Windsurf more reliable?

Based on StupidLLM data, Claude Code has 8 documented failures (avg severity 6.4/10) while Windsurf has 1 (avg severity 7.5/10). Claude Code shows better reliability based on average severity scores.

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

Claude Code's most common failure mode is destructive action, while Windsurf most commonly fails via ignored instructions. Claude Code has 3 critical incidents vs Windsurf's 0.