Count documents in a collection
AI agents call count_documents to retrieve information from Firestore MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Counting documents is a read-only operation that queries collection metadata and returns a number. It has no side effects, does not modify data, does not execute arbitrary code, and does not delete or commit financial transactions. The blast radius of misuse is minimal — an agent could only retrieve count information, which is informational metadata.
From the tool's definition Tool name is 'count_documents' and description states 'Count documents in a collection' — this is a query operation that retrieves aggregate information without modifying, deleting, or executing code.
Attacks that exploit this kind of access
Count documents in a collection. It is categorised as a Read tool in the Firestore MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Firestore MCP Server MCP server in PolicyLayer and add a rule for count_documents: 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 Firestore MCP Server. Nothing to install.
count_documents 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_documents 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_documents. 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_documents is provided by the Firestore MCP Server MCP server (yudai-uk/firestore-mcp). 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 →