AI agents call connections_graph to retrieve information from Arkime without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves network session metadata to construct a graph visualization. It is a read-only operation that extracts information from the Arkime packet capture database for analysis purposes. No data is modified, deleted, or used to trigger external actions. It aligns with the Read category as a data retrieval function with no side effects.
From the tool's definition The tool description states 'Get network connection graph — nodes and links between sources and destinations.' The verb 'Get' indicates data retrieval. No create, modify, delete, or execute operations are described.
Attacks that exploit this kind of access
Get network connection graph — nodes and links between sources and destinations. It is categorised as a Read tool in the Arkime MCP Server, which means it retrieves data without modifying state.
Register the Arkime MCP server in PolicyLayer and add a rule for connections_graph: 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 Arkime. Nothing to install.
connections_graph 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 connections_graph 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 connections_graph. 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.
connections_graph is provided by the Arkime MCP server (swannman/arkime-mcp-server). 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 →