self-learning

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly instructs the agent to "use web search" and "fetch and extract content from the top 3-5 URLs found" and lists "Web page fetching capability" as a required tool, meaning it will ingest open/public third‑party webpages (e.g., documentation, tutorials, gists) and use their contents to drive research, generation, and saving actions—allowing untrusted third‑party content to influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly fetches and extracts content from the "top 3-5 URLs" found via web search at runtime (i.e., arbitrary external URLs discovered and fetched during Step 3), and that fetched content is used to directly control the generated SKILL.md/prompts, so any such externally fetched URL (the runtime "top 3-5 URLs" results) is a risky dependency.