AI agents call count_reports to retrieve information from Humaans without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a query operation that retrieves or calculates reporting structure information without altering any state. It has no destructive, financial, or code-execution capabilities. The 'Read-only' server designation and the tool's counting/querying nature confirm the Read category. Severity is low because exposure of this tool to an AI agent poses minimal risk—counting report hierarchies cannot cause harm.
From the tool's definition Tool description states it 'count[s] reports for a person' with options for direct vs. recursive counts. Server is explicitly described as 'Read-only MCP server' providing tools to 'query' data.
Attacks that exploit this kind of access
Count reports for a person. recursive=False gives direct reports only; True gives the full subtree size. It is categorised as a Read tool in the Humaans MCP Server, which means it retrieves data without modifying state.
Register the Humaans MCP server in PolicyLayer and add a rule for count_reports: 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 Humaans. Nothing to install.
count_reports 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_reports 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_reports. 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_reports is provided by the Humaans MCP server (ptorsten/humaans-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 →