AI agents invoke cmux_send_panel to trigger actions in Cmux. 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.
Sending text into a terminal panel effectively executes arbitrary commands or input in that panel. The blast radius is high because an AI agent could send any shell command, script, or destructive input to an active terminal session in any workspace.
From the tool's definition 'Send text into an agent panel via send-panel' — sends arbitrary text/commands into a terminal pane, which executes whatever input is sent
Attacks that exploit this kind of access
Send text into an agent panel via send-panel. Use cmux_list_panels to find the panel ref. Pass workspace/window when the panel lives in another workspace. It is categorised as a Execute tool in the Cmux MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Cmux MCP server in PolicyLayer and add a rule for cmux_send_panel: 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. Nothing to install.
cmux_send_panel 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_send_panel 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_send_panel. 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_send_panel is provided by the Cmux MCP server (puchkoff/cmux-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 →