skill-docs

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

• Third-party content exposure detected (high risk: 0.95). The required workflow uses WebFetch to fetch and extract full content from an outsider-authored public web page at runtime (“https://code.claude.com/docs/en/skills”), which becomes readable text in the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The SKILL explicitly instructs the agent to use the WebFetch tool at runtime to fetch and "extract the full content" from https://code.claude.com/docs/en/skills and "apply it" to the current task, meaning remote content would be fetched during runtime and directly used to control the agent's instructions and behavior.