MCP Guardrail Server

MCPOpen SourceMIT16.0

by atamaplus-public • Security & Privacy

A secure MCP server that executes only pre-authorized shell commands, providing controlled command execution for AI assistants in high-security environments.

Example Use Cases

1
Run a limited set of shell commands (e.g., git, ls, npm) without exposing full system access.
2
Integrate safe command execution into Claude Desktop or similar MCP consumers while enforcing an allowlist and timeouts.
3
A simple, auditable way to test or demo command execution workflows in development or high-security environments.

Description

MCP Guardrail Server runs a minimal MCP-compatible server that allows execution only of commands on a configured allowlist and enforces execution timeouts. It is suitable for restricting an agent's capabilities to a safe subset of system commands. The project includes integration instructions for Claude Desktop, test scripts to verify allowed/blocked commands, and is distributed with an MIT license.

Quick Actions

View on GitHub

Security

Scanned 2 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

D44/100

4/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities0 Tools / 0 Prompts / 0 Resources

Owneratamaplus-public

CategorySecurity & Privacy

DependenciesStandalone