plan

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md:14-15 requires reading “every file in requirement_dir” and then using that content in “Gather context” (SKILL.md:21-23), so if those requirement files are authored by outsiders, their free-form text is ingested into the agent’s LLM context at runtime.