Count documents in a collection matching a query
AI agents call count to retrieve information from MongoDB without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The count tool performs a read-only operation that retrieves statistical information about documents matching a query filter. It has no side effects, does not modify data, and cannot be used to create, update, delete, or execute arbitrary code. It is a simple aggregation query that falls squarely into the Read category.
From the tool's definition Tool is described as 'Count documents in a collection matching a query' and the server description explicitly states it 'provides access to MongoDB databases' for 'read-only queries.' The count operation is a query that retrieves aggregate information without…
Documented attack patterns abuse exactly the kind of access count gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MongoDB, and nothing reaches the server without passing your rules. This is the rule we recommend for count:
{
"version": "1",
"default": "deny",
"tools": {
"count": {}
}
}count is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Count documents in a collection matching a query. It is categorised as a Read tool in the MongoDB MCP Server, which means it retrieves data without modifying state.
Register the MongoDB MCP server in PolicyLayer and add a rule for count: 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 MongoDB. Nothing to install.
count 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 count 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 count. 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.
count is provided by the MongoDB MCP server (kiliczsh/mcp-mongo-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MongoDB, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
8 MongoDB tools catalogued and risk-classified — across an index of 43,000+ MCP servers.