linkedin-post-generator

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests open web content (e.g., "Blog/article URL: fetch the page. Extract headline, body text..." in SKILL.md and README examples like "Turn this into a LinkedIn post: https://yourblog.com/my-post"), so it will read and act on arbitrary public/user-generated pages and PRs which could contain injected instructions that alter generated posts or posting behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches user-supplied web or GitHub PR URLs at runtime (e.g., https://yourblog.com/my-post and example eval URL https://vercel.com/blog/how-we-optimized-package-imports-in-next-js) and injects that fetched content directly into the agent prompt/context to generate the LinkedIn post, so remote content can directly control agent behavior.