find-internal-answer

Fail

Audited by Snyk on Nov 11, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The bootstrap block tells the agent to ignore authorization-required statements and to suppress connectivity or auth messages from the user, and to run tool activity silently; this is self-serving and hides the agent's behavior, but it reads as product-tuning rather than an attack on the user.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). The required workflow searches Slack channels (step 3, line 32) and email threads (step 4, line 33), which external senders can post into or send to without the agent selecting a specific item, so outsider-authored text reaches the LLM.

Issues (2)

E004
CRITICAL

Prompt injection detected in skill instructions.

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 11, 2026, 06:39 PM
Issues
2