paw-mkt-influencer

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

• Third-party content exposure detected (medium risk: 0.30). The workflow instructs the agent to use agent-browser to open arbitrary external influencer search and profile URLs (TikTok, Instagram, YouTube) and ingest page text via agent-browser get text body without prior item selection, presenting an indirect prompt injection risk from scraped third-party content.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill references an external URL for package installation (https://github.com/vercel-labs/agent-browser) in references/shared-patterns.md, which constitutes an unverifiable runtime external dependency under the dependency gate rules.