AI agents call youtube_search to retrieve information from Google without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
YouTube search is a standard information retrieval operation with no capability to create, modify, delete, or execute external actions. The blast radius of misuse is minimal—an agent could search for inappropriate content or spam queries, but this does not compromise data integrity, financial systems, or enable destructive actions. This clearly falls under the Read category.
From the tool's definition Tool name is 'youtube_search' and description states 'Search YouTube videos by query.' This is a read-only operation that retrieves and queries data without modification or side effects.
Attacks that exploit this kind of access
Search YouTube videos by query. It is categorised as a Read tool in the Google MCP Server, which means it retrieves data without modifying state.
Register the Google MCP server in PolicyLayer and add a rule for youtube_search: 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 Google. Nothing to install.
youtube_search 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 youtube_search 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 youtube_search. 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.
youtube_search is provided by the Google MCP server (psckeithw/mcp-google). 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 →