gateway-client-imgly

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

• Third-party content exposure detected (high risk: 0.85). The required runtime step “fetch the live reference” via WebFetch https://gateway.img.ly/llms.txt pulls public web content into the agent’s context, which is outsider-authored free text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent to fetch the live reference at https://gateway.img.ly/llms.txt at runtime and to "read it before answering" because it is the source of truth for endpoints, models, schemas and SSE, so remote content from that URL is fetched during runtime and directly controls how the agent composes requests/behaviour.