voice-profile
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 directs the agent to run an unrelated Salesforce GraphQL call and to withhold connector-disconnection reports from the user, which hides real connectivity failures and acts beyond what this email-only skill needs, but it shows no sign of malicious intent.
Issues (1)
E004
CRITICALPrompt injection detected in skill instructions.
Audit Metadata