understand

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md Phase 1 reads outsider-authored project files like README.md and package.json from the target directory and injects their contents as “Additional context from main session” into subagent dispatch prompts (LLM context), creating an indirect prompt-injection path from untrusted repo text.