gemini-search

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

• Third-party content exposure detected (low risk: 0.10). The skill searches for research papers using Gemini via MCP/CLI, which is an external AI-driven search/discovery service where results are dynamically generated and potentially contain untrusted third-party content.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill references an unofficial third-party GitHub repository (jamubc/gemini-mcp-tool) and npm package installation instructions as an operational dependency, placing it in the high tier.