AI agents call aggregate_flows to retrieve information from Tshark without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Despite the uninformative empty description, the verb 'aggregate' typically indicates summarization or grouping of already-captured packet/flow data without modification. The sibling tools are predominantly Read (analyze_*, extract_*) or Execute (capture_*, export_*) in nature. Given TShark's nature as a packet analyzer, aggregating flows would retrieve and compute statistics over flows without side effects.
From the tool's definition Tool name 'aggregate_flows' suggests computation/summarization of network flow data. The empty description provides no direct evidence, but the name pattern and context (network packet analysis suite) implies read-only aggregation of existing capture data.
Attacks that exploit this kind of access
aggregate_flows. It is categorised as a Read tool in the Tshark MCP Server, which means it retrieves data without modifying state.
Register the Tshark MCP server in PolicyLayer and add a rule for aggregate_flows: 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 Tshark. Nothing to install.
aggregate_flows 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 aggregate_flows 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 aggregate_flows. 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.
aggregate_flows is provided by the Tshark MCP server (ouonet/tshark-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 →