Close the browser and clean up the MCP connection to Google Flow.
AI agents invoke flow_disconnect to trigger actions in Google Flow Browser MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool terminates an active browser session and cleans up the MCP connection. It executes a teardown operation that ends external browser automation and connection state. While not permanently destructive to data, it disrupts ongoing operations and cannot be trivially undone without reconnecting, placing it in Execute rather than Destructive (no data is deleted).
From the tool's definition 'Close the browser and clean up the MCP connection to Google Flow'
Documented attack patterns abuse exactly the kind of access flow_disconnect gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Google Flow Browser MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for flow_disconnect:
{
"version": "1",
"default": "deny",
"tools": {
"flow_disconnect": {
"limits": [
{
"counter": "flow_disconnect_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}flow_disconnect stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Close the browser and clean up the MCP connection to Google Flow. It is categorised as a Execute tool in the Google Flow Browser MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Google Flow Browser MCP server in PolicyLayer and add a rule for flow_disconnect: 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 Google Flow Browser MCP. Nothing to install.
flow_disconnect is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the flow_disconnect 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 flow_disconnect. 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.
flow_disconnect is provided by the Google Flow Browser MCP server (tmsss05/google-flow-browser-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Google Flow Browser MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
17 Google Flow Browser MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.