keyapi-instagram

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md’s required workflow includes resolving KeyAPI endpoint docs by fetching https://docs.keyapi.ai/llms.txt and linked endpoint pages at runtime (public web content), which is then parsed and used to build the LLM-visible execution context (e.g., method/path/params/body/response contract).

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's search-keyapi-docs.mjs fetches https://docs.keyapi.ai/llms.txt (and may fetch linked https://docs.keyapi.ai/en/instagram/... pages) at runtime to extract OpenAPI methods/paths/params which directly control which API calls the agent constructs and executes, and the SKILL mandates this docs resolution as a required step.