How does Dynamic ensure secure execution of tools?

Dynamic isolates each tool execution in a hardened Docker container. This includes read-only file systems, dropped capabilities, and strict resource limits (PID, memory, CPU) to prevent abuse and enhance security.

Can I manage tools without restarting the Dynamic server?

Yes, Dynamic features a robust control plane (`dynamic.tool.*`) that allows AI agents and operators to create, update, delete, enable, and disable tools dynamically without requiring any server restarts.

Is Dynamic suitable for production environments?

Absolutely. Dynamic is production-ready with features like health probes, Prometheus metrics, graceful shutdown, Kubernetes manifests, dual registry backends (File/PostgreSQL), and optional JWT authentication for secure deployments.

What is Dynamic and its core capability?

Dynamic is a Node.js-based MCP server that empowers AI agents and operators to dynamically create, manage, and execute tools at runtime within secure, isolated Docker sandboxes.

Dynamic

Name: Dynamic
Author: mcpland

bymcpland

•

Developer Tools

Security & Testing

Deployment & DevOps

Enables runtime creation, management, and execution of tools for AI agents and operators in isolated Docker sandboxes.

Dynamic is a robust Node.js server designed to overcome the limitations of static MCP servers. It empowers AI agents and operators to dynamically create, update, and delete tools on the fly, with each tool running in a hardened Docker container. It offers comprehensive lifecycle management, dual transport options (stdio/HTTP), flexible registry backends (file/PostgreSQL), and essential production features like JWT authentication, audit logging, and Prometheus metrics, ensuring a secure and scalable environment for dynamic tool execution.

Key Features

01Runtime tool management via control plane (create, update, delete, enable/disable tools without restarts)

02Docker sandbox isolation for secure tool execution with resource limits and dropped capabilities

03Dual transport (Stdio for local use, Streamable HTTP for networked deployments) and dual registry backend (File/PostgreSQL)

04Execution guard for global concurrency and per-scope rate limiting to prevent abuse

05Production-ready with health probes, Prometheus metrics, graceful shutdown, and Kubernetes manifests

061 GitHub stars

Use Cases

01Integrating dynamic tool creation and management with AI agents (e.g., Claude Desktop, Claude Code)

02Deploying a scalable and secure MCP server in production environments using Docker or Kubernetes

03Executing one-off JavaScript or shell commands in isolated sandboxes through the server