Get the raw output buffer from a session.
AI agents call terminal_get_buffer to retrieve information from Mcp Shellkeeper without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves historical output/state from an already-established terminal session. It performs no code execution, data modification, or destructive operations. The 'Get' verb and 'raw output buffer' retrieval confirm it is a read-only operation with minimal security impact unless the buffer contains sensitive credentials or secrets.
From the tool's definition Tool name 'terminal_get_buffer' and description 'Get the raw output buffer from a session' indicate retrieval of existing session output without modification or execution.
Attacks that exploit this kind of access
Get the raw output buffer from a session. It is categorised as a Read tool in the Mcp Shellkeeper MCP Server, which means it retrieves data without modifying state.
Register the Mcp Shellkeeper MCP server in PolicyLayer and add a rule for terminal_get_buffer: 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 Mcp Shellkeeper. Nothing to install.
terminal_get_buffer 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 terminal_get_buffer 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 terminal_get_buffer. 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.
terminal_get_buffer is provided by the Mcp Shellkeeper MCP server (mcp-shellkeeper). 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 →