Execute multiple CMUX operations in a single MCP call. Each step references an existing cmux tool by name and provides its parameters. Steps run sequentially. Later steps can reference outputs from earlier steps using $steps[N].path.to.field syntax in string parameter values. Example: Launch agen...
AI agents invoke cmux_batch to trigger actions in Cmux Agent. 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 orchestrates sequential execution of multiple CMUX operations, including shell commands and browser interactions. The ability to execute arbitrary sequences of operations from the 80+ available tools (evidenced by siblings like cmux_browser_eval and cmux_browser_navigate), combined with variable substitution and chaining, makes this an Execute category tool.
From the tool's definition Execute multiple CMUX operations in a single MCP call. Steps run sequentially. Later steps can reference outputs from earlier steps using $steps[N].path.to.field syntax.
Documented attack patterns abuse exactly the kind of access cmux_batch gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Cmux Agent, and nothing reaches the server without passing your rules. This is the rule we recommend for cmux_batch:
{
"version": "1",
"default": "deny",
"tools": {
"cmux_batch": {
"limits": [
{
"counter": "cmux_batch_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}cmux_batch 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.
Execute multiple CMUX operations in a single MCP call. Each step references an existing cmux tool by name and provides its parameters. Steps run sequentially. Later steps can reference outputs from earlier steps using $steps[N].path.to.field syntax in string parameter values. Example: Launch agents, set progress, and orchestrate — all in one call instead of 8+ separate tool calls. Variable substitution: In any string param value, use $steps[0].surfaces[2] to reference the 3rd surface from step 0. It is categorised as a Execute tool in the Cmux Agent MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Cmux Agent MCP server in PolicyLayer and add a rule for cmux_batch: 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 Cmux Agent. Nothing to install.
cmux_batch 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 cmux_batch 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 cmux_batch. 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.
cmux_batch is provided by the Cmux Agent MCP server (multiagentcognition/cmux-agent-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Cmux Agent, 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.
63 Cmux Agent tools catalogued and risk-classified — across an index of 43,000+ MCP servers.