explain-code

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

• Third-party content exposure detected (high risk: 0.75). The required workflow instructs the agent to “Search the web” and “WebFetch” for up-to-date docs (SKILL.md:33-37), which can ingest outsider-authored free text from public web pages into the LLM context at runtime.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs runtime fetching of live documentation via WebSearch/WebFetch and the GitHub API (e.g., gh api repos/{owner}/{repo}/releases/latest), which would inject external content into the model context and thus directly control prompts at runtime.