alterlab-bgpt-search

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

• Third-party content exposure detected (low risk: 0.10). The agent retrieves search results and scientific paper content from the BGPT MCP service, which contains third-party authored publication data.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to connect to an external remote Model Context Protocol (MCP) server at bgpt.pro, an arbitrary third-party service whose publisher provenance and trustworthiness are unverified.