Search the web using Gemini with Google Search grounding.
AI agents call gemini_search_tool to retrieve information from MCP Gemini without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs web search, which is a retrieval operation with no side effects. It queries data and returns results without creating, modifying, deleting, or executing code. The grounding with Google Search indicates it's a bounded read operation. Severity is low because misuse would result in unwanted search results or information disclosure, not data loss or code execution.
From the tool's definition Tool description states 'Search the web using Gemini with Google Search grounding.' Search operations are read-only queries that retrieve information without modifying or executing external operations.
Attacks that exploit this kind of access
Search the web using Gemini with Google Search grounding. It is categorised as a Read tool in the MCP Gemini MCP Server, which means it retrieves data without modifying state.
Register the MCP Gemini MCP server in PolicyLayer and add a rule for gemini_search_tool: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches MCP Gemini. Nothing to install.
gemini_search_tool is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the gemini_search_tool rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for gemini_search_tool. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
gemini_search_tool is provided by the MCP Gemini MCP server (mcp-gemini-crunchtools). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →