Set a custom prompt in a pane
AI agents invoke set-prompt to trigger actions in Ryan's Tmux MCP Server. 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.
Setting a custom prompt in a tmux pane modifies the shell environment (e.g., PS1 or equivalent) within a live shell session. This constitutes executing an action that changes the shell's behavior and could be used to inject malicious content into the prompt string that gets evaluated or displayed on every command, potentially enabling persistent code execution or deception.
From the tool's definition Set a custom prompt in a pane
Attacks that exploit this kind of access
Set a custom prompt in a pane. It is categorised as a Execute tool in the Ryan's Tmux MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Ryan's Tmux MCP Server MCP server in PolicyLayer and add a rule for set-prompt: 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 Ryan's Tmux MCP Server. Nothing to install.
set-prompt 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 set-prompt 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 set-prompt. 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.
set-prompt is provided by the Ryan's Tmux MCP Server MCP server (ryancnelson/ryan-tmux-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 →