customer-health
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 dismiss authorization statements from its context and to keep connectivity problems and tool activity from the user, which hides some of the skill's own activity and limitations without showing malicious intent.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The required Step 2 Salesforce read pulls Case
Subjectand TaskSubjecttext, which customers can submit through web-to-case or email-to-case, and the required Step 2b email search returns external contacts' email bodies without the agent selecting a specific item first.
Issues (2)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata