sre
The SmythOS Runtime Environment (SRE) is an open-source, cloud-native runtime for agentic AI. Secure, modular, and production-ready, it lets developers build, run, and manage intelligent agents across local, cloud, and edge environments.
claude mcp add --transport stdio smythos-sre node packages/core/dist/index.js
How to use
SmythOS SRE is the Smyth Runtime Environment, an open-source kernel for AI agents, together with the SDK and CLI that let you build, run, and manage agents in a unified way. The system provides OS-level abstractions for resources like large language models, vector databases, storage, and caches, plus an extensible component library for production-grade agent orchestration and security. Use the CLI to scaffold projects, leverage the SDK to write agent logic in TypeScript, and rely on the core runtime (SRE) to execute agents consistently across environments. The available tooling includes: the SRE CLI for project creation and management, the SDK for building agent code with type safety and a unified API, and the core runtime that exposes the resource abstractions and agent lifecycle management. You can explore examples under the repository and the online documentation for SDK usage, core concepts, and integration patterns.
How to install
Prerequisites:
- Node.js (LTS) and npm installed on your system
- Git to clone or fetch the repository
Installation steps:
- Install the CLI globally to scaffold and manage SmythOS projects:
npm i -g @smythos/cli
- Create a new SmythOS project (through the CLI):
sre create
- Install the SDK into an existing project if you prefer direct SDK usage:
npm install @smythos/sdk
- (Optional) Install and run the core runtime locally if you want to experiment with the server directly. The entry point used by MCP is the core runtime script; ensure Node can resolve packages/core accordingly:
node packages/core/dist/index.js
- Refer to the repository’s examples and documentation for environment-specific deployment instructions and configuration details.
Note: If you run into CLI or code issues, set LOG_LEVEL=debug to collect detailed logs for troubleshooting.
Additional notes
Tips and common issues:
- Ensure Node.js version aligns with the project’s requirements (LTS recommended).
- Use the CLI’s debug mode (LOG_LEVEL=debug) to diagnose problems during project creation or agent execution.
- The unified resource abstraction covers storage, LLMs, vector databases, and caches; when swapping providers, keep the same API usage to minimize code changes.
- Review the SDK typings for strong type safety and autocompletion in IDEs.
- Check the repository’s examples and the online docs for integration templates and code templates to accelerate development.
Related MCP Servers
voltagent
AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework
DeepMCPAgent
Model-agnostic plug-n-play LangChain/LangGraph agents powered entirely by MCP tools over HTTP/SSE.
sdk-typescript
A model-driven approach to building AI agents in just a few lines of code.
Autono
A ReAct-Based Highly Robust Autonomous Agent Framework.
weam
Web app for teams of 20+ members. In-built connections to major LLMs via API. Share chats, prompts, and agents in team or private folders. Modern, fully responsive stack (Next.js, Node.js). Deploy your own vibe-coded AI apps, agents, or workflows—or use ready-made solutions from the library.
c4-genai-suite
c4 GenAI Suite