lib-docs

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly instructs the agent to WebSearch and WebFetch external documentation (and references/sources.md lists public sites such as Stack Overflow, GitHub, and various public docs URLs), so the agent ingests and acts on untrusted third‑party content.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly performs WebFetch at runtime to pull documentation pages (e.g., https://github.com/seancorfield/honeysql#getting-started), and those fetched pages are injected/used to determine the agent’s responses, meaning the external content can directly control prompts.