amazon-seller-lead-research

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill instructs the agent to automatically download and install a runtime dependency without user confirmation.

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

• Third-party content exposure detected (low risk: 0.10). The skill browses external web pages and merchant profiles via BrowserAct where outsider-authored content can be ingested.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to automatically download and install a runtime skill/dependency from an external GitHub repository via an arbitrary web redirect URL, presenting a weak-provenance operational dependency.