Model Context Protocol moderate risk

Term MCP DeepSeek

A DeepSeek-powered MCP-like server enabling AI agents to interact with a terminal session, execute commands, and manage sessions via HTTP or STDIO.

Connections & Capabilities

Connects To

GitHubDocker

Capabilities

readwriteexec

Quickstart

Install

pip install -r

Exposed MCP Tools (3)

high

write_to_terminal

Writes a command to the active terminal session.

Allows arbitrary command execution on the server.

safe

read_terminal_output

Retrieves the output from the active terminal session.

Read-only operation with no side effects.

moderate

send_control_character

Sends a control character to the terminal session (e.g., Ctrl+C).

Can interrupt or terminate processes, potentially causing disruption.

Safety Assessment

This server allows AI agents to execute shell commands, which inherently carries risk. While it includes some security measures, it's best used in controlled environments with trusted AI agents. Avoid exposing it directly to the public internet without additional security layers.

Basic authentication implemented.
Rate limiting to prevent abuse.
Input validation to mitigate command injection.
Security headers to enhance security posture.

Executes shell commands, posing a risk of arbitrary code execution.
Relies on environment variables for API key, which can be insecure if not managed properly.
Basic authentication may not be sufficient for high-security environments.
Lacks comprehensive error handling, potentially exposing vulnerabilities.