AgentProv
← All incidents
2025-08 · 2025-Q3

Claude Code InversePrompt Command Injection

Layer tags: not yet tagged.

Target

Anthropic Claude Code (below v1.0.20)

Impact

Whitelisted echo command used as injection vector; AI model helps reverse-engineer its own security

Root Cause

Error in command parsing; echo whitelisted without sanitization

View this entry in the catalog