AI agents invoke unitree_connect to trigger actions in Robot. 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.
Based on the server context (controlling Unitree robots with mock and hardware backends) and the tool name, this likely establishes a connection to a physical robot. Initiating a connection to hardware can trigger real-world effects and is analogous to an Execute action. Severity is high given that connecting to a physical robot is a precursor to potentially dangerous physical operations.
From the tool's definition Tool name 'unitree_connect' on a robot-control server; description is empty and uninformative.
Documented attack patterns abuse exactly the kind of access unitree_connect gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Robot, and nothing reaches the server without passing your rules. This is the rule we recommend for unitree_connect:
{
"version": "1",
"default": "deny",
"tools": {
"unitree_connect": {
"limits": [
{
"counter": "unitree_connect_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}unitree_connect 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.
unitree_connect. It is categorised as a Execute tool in the Robot MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Robot MCP server in PolicyLayer and add a rule for unitree_connect: 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 Robot. Nothing to install.
unitree_connect 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 unitree_connect 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 unitree_connect. 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.
unitree_connect is provided by the Robot MCP server (showkeyjar/robot-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Robot, 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.
24 Robot tools catalogued and risk-classified — across an index of 43,000+ MCP servers.