skill-factory

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

• Third-party content exposure detected (high risk: 0.90). Yes — the skill explicitly automates scraping/ingestion of public documentation, GitHub repos, PDFs and arbitrary URLs via Skill_Seekers (see SKILL.md and references/skill-seekers-integration.md and references/request-analysis.md), and then reads and acts on that scraped content in its automatic creation/QA/testing workflows, so untrusted third‑party content can materially influence tool use and agent actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill auto-installs and runs Skill_Seekers by cloning and executing code from https://github.com/yusufkaraaslan/Skill_Seekers and uses it at runtime to scrape arbitrary documentation sites (e.g., https://react.dev, https://docs.rs) whose fetched content is transformed into SKILL.md and thus directly controls agent instructions — a clear runtime external-code + prompt-injection dependency.