brainstorming

Fail

Audited by Snyk on May 6, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (high risk: 0.80). The prompt explicitly instructs the brainstorming skill to auto-continuously invoke the implementation-planning skill (/exarchos:plan) and perform state/side-effecting operations (auto-chain, orchestration calls, file saves) despite the header saying "Do NOT use for implementation planning", so it contains deceptive behavior that goes beyond the skill's stated purpose.

Issues (1)

E004
CRITICAL

Prompt injection detected in skill instructions.

Audit Metadata
Risk Level
CRITICAL
Analyzed
May 6, 2026, 04:03 AM
Issues
1
Security Audit — snyk — brainstorming