marketing-launch

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md “Research Mode: Launch Intelligence,” the workflow uses agent-browser to open and get text body from outsider-authored pages on Product Hunt, competitor blogs/social (generic URLs), Reddit search results, LinkedIn posts, and Indie Hackers, ingesting their free text at runtime without selecting specific items.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's shared-patterns setup instructs installing agent-browser via "npx skills add https://github.com/vercel-labs/agent-browser", which fetches and executes remote code at runtime and is relied upon for Research Mode, so the GitHub URL is a runtime external dependency that can execute code.