Get current terminal output. Defaults to the
AI agents call get_output to retrieve information from Persistent Shell MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Although the server enables command execution (which would be Execute/Destructive), this specific tool only retrieves output from existing processes. It does not execute commands, modify data, delete resources, or commit financial actions. The blast radius of misuse is minimal—an AI agent could only read information already displayed in terminal sessions, not cause harm through this tool alone.
From the tool's definition Tool name is 'get_output' and description indicates it retrieves current terminal output with no modification capability ('Get current terminal output. Defaults to the...'). This is a read-only operation that queries state without side effects.
Attacks that exploit this kind of access
Get current terminal output. Defaults to the. It is categorised as a Read tool in the Persistent Shell MCP MCP Server, which means it retrieves data without modifying state.
Register the Persistent Shell MCP server in PolicyLayer and add a rule for get_output: 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 Persistent Shell MCP. Nothing to install.
get_output is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_output 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 get_output. 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.
get_output is provided by the Persistent Shell MCP server (tntisdial/persistent-shell-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 →