url-to-code

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

• Third-party content exposure detected (high risk: 0.85). The workflow explicitly opens a user-supplied source URL at runtime (SKILL.md:24-30) and then “capture[s] the source page carefully” and uses browser DOM tools to gather elements/text/links/styles (SKILL.md:32-60), which would ingest outsider-authored free text from the fetched webpage into the agent’s LLM context.