paw-mkt-pricing

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

• Third-party content exposure detected (low risk: 0.10). The agent skill instructs the LLM to scrape live competitor pricing pages and retrieve body text using browser automation (agent-browser get text body and WebFetch), which exposes the agent to arbitrary outsider-authored web content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The instruction to install agent-browser via npx skills add https://github.com/vercel-labs/agent-browser --skill agent-browser points to a third-party GitHub repository, constituting an unverified operational external dependency under the dependency gate.