AI agents call web_search_tool to retrieve information from Web without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Web search is a read-only operation that queries external data sources and returns information. It does not create, modify, delete, or execute code. The blast radius of misuse is minimal — worst case, an agent could search for sensitive information, but this requires the information to already exist publicly and searchable online. No system resources, data stores, or external services are materially altered.
From the tool's definition Tool description states 'Search the web using DuckDuckGo and return results' — a query operation that retrieves data without modification or side effects.
Attacks that exploit this kind of access
Search the web using DuckDuckGo and return results. It is categorised as a Read tool in the Web MCP Server, which means it retrieves data without modifying state.
Register the Web MCP server in PolicyLayer and add a rule for web_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 Web. Nothing to install.
web_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 web_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 web_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.
web_search_tool is provided by the Web MCP server (ryandam/essentialmcp). 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 →