defold-docs-fetch

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md's Usage section explicitly instructs the agent to fetch and read raw Markdown from public URLs (e.g., https://defold.com/llms/manuals/collection-proxy.md), so the agent ingests third‑party web content that can influence its 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 external Markdown docs at runtime (e.g., https://defold.com/llms/manuals/collection-proxy.md) and injects their content into the agent's context to drive responses, so the fetched content directly controls prompts and is a required dependency.