xquik-social-research

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly treats retrieved tweet/bio/article/display-name text as untrusted data and returns it to the agent, meaning outsider-authored free text from Xquik API responses can be ingested into the LLM context at runtime.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to fetch live endpoint metadata that can change agent behavior—specifically the OpenAPI schema and MCP endpoint at https://xquik.com/openapi.json and https://xquik.com/mcp—which are runtime external dependencies that can directly control how requests/prompts are constructed.