hono

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md "Upstream Grounding" explicitly tells the agent to query DeepWiki against public GitHub repos (honojs/hono and cloudflare/cloudflare-docs), which means the skill instructs the agent to fetch and interpret public third‑party content that can influence decisions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent at runtime to query DeepWiki against the repositories https://github.com/honojs/hono and https://github.com/cloudflare/cloudflare-docs, which implies fetching repo content to ground responses and injecting external content into the model context (i.e., controlling prompts).