Vendors and modelsModelby Anthropic

Claude Code AI failures

Every documented AI failure involving Claude Code on the AI Failure Index, classified by the mechanism that broke.

Failures
1
Highest severity
High
Span
2025
Failure modes
1
By failure mode
By severity
FI-0029SaaSHigh
Agentic Action Error

Claude Code ran rm -rf on a user's home directory while rebuilding a project

A developer asked Anthropic's Claude Code to rebuild a Makefile project from a fresh checkout. The agent generated and executed a command whose trailing path expanded to the user's full home directory, deleting years of files. He was not running with the skip-permissions flag.

Confidence
High (multi-source, primary)
Anthropic (Claude Code)2 sourcesPrimaryPublicOct 2025

See how Realm catches these failure modes at runtime.

Explore Realm Labs