AI agents call get_flows to retrieve information from Mitm without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves previously captured network traffic flows. While it performs no write, delete, or execution action, it accesses sensitive intercepted data including credentials, API keys, and TLS handshakes. This makes it a Read operation with medium severity due to the sensitive nature of the data exposed (passwords, tokens, certificate details), but without the blast radius of Execute or Destructive categories.
From the tool's definition Tool name 'get_flows' and description 'Read intercepted HTTP/TLS flows from a session, with optional filtering' explicitly states it retrieves data without side effects.
Attacks that exploit this kind of access
Read intercepted HTTP/TLS flows from a session, with optional filtering. It is categorised as a Read tool in the Mitm MCP Server, which means it retrieves data without modifying state.
Register the Mitm MCP server in PolicyLayer and add a rule for get_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 Mitm. Nothing to install.
get_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 get_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 get_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.
get_flows is provided by the Mitm MCP server (mplogas/mitmproxy-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 →