http-request

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill accepts arbitrary headers and bodies (e.g., Authorization headers or API keys) and thus an agent invoking it would need to insert secret values verbatim into the request payload, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 1.00). The skill accepts an arbitrary URL in its required input (input_schema "url") and, as shown in runtime/network settings and scripts/main.py, fetches open web pages (examples: Wikipedia, wttr.in) and converts HTML to Markdown for the agent to read, thereby exposing the agent to untrusted third-party content that could contain instructions.