SmythOS/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.

What it solves

SmythOS SRE provides a standardized infrastructure layer for building and deploying production-ready AI agents. It eliminates the need to write custom glue code for different AI services by providing a unified API that works across various LLM, vector database, storage, and caching providers.

How it works

Inspired by operating system kernels, the project consists of three main parts:

  • SRE (Smyth Runtime Environment): The core "kernel" that manages AI resources through a modular architecture of pluggable connectors (e.g., OpenAI for LLMs, Pinecone for VectorDBs, S3 for storage).
  • SDK: A TypeScript-based abstraction layer that allows developers to write agent logic once and scale it across different infrastructure providers without changing the business logic.
  • CLI: A command-line tool for project scaffolding and management.

Who it’s for

Developers and enterprise teams who want to build scalable, secure AI agents and need a way to swap infrastructure providers (like switching from local storage to S3 or OpenAI to Anthropic) without rewriting their code.

Highlights

  • Unified Resource Abstraction: Consistent APIs for LLMs, VectorDBs, Cache, and Storage regardless of the provider.
  • Smyth File Support: Ability to load and run agents defined in .smyth files created via a visual builder.
  • Enterprise Security: Built-in access control lists (ACL) and secure credential management via a Candidate/ACL system.
  • Extensive Component Library: Over 40 production-ready components for AI, data processing, and logic (e.g., WebSearch, JSONFilter, ImageGen).