web-exploit-technique

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

• Third-party content exposure detected (low risk: 0.10). The required workflow fetches and reads responses from a user-supplied target web application (HTTP responses, page source, CSP headers, and exposed files), which an outsider could influence only after the agent selects that target, so this is low exposure.

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

• Potentially malicious external URL detected (medium risk: 0.30). The only inbound operational reference is the Gopherus GitHub repository, which the SSRF reference instructs the agent to use by running its gopherus.py code, and that individual-owned repository is unofficial with no concrete warning signs, so it is a common/reputable unofficial dependency (medium).