Skip to main content
Asterai is the tool marketplace for AI agents. Discover, share, and deploy tools written in any programming language, with instant cloud execution.

​
What is asterai?

Asterai is a registry and runtime for portable, sandboxed tools. Write a tool once in your preferred language, publish it to the registry, and run it anywhere—locally or in the cloud. Under the hood, asterai uses WebAssembly for portability and security. But you don’t need to know anything about WebAssembly to use asterai. Just write code in a supported language (TypeScript, Python, Rust, Go, and more), and asterai handles the rest.

​
Key Concepts

  1. Component: A portable tool or library. Components are published to the registry and can be composed together. Write them in any supported language.
  2. Environment: A deployable bundle of one or more components with configuration (environment variables, secrets). Environments can run locally or in the cloud.
  3. Registry: Where components are published and discovered. Each user has a namespace for their components.
  4. CLI: The open-source command-line tool for building, publishing, and running components locally.
  5. Cloud: Run environments on asterai’s infrastructure with a simple API call—no DevOps required.

​
Why asterai?

Language interoperability Write tools in any language. Components written in different languages work together seamlessly through typed interfaces. Sandboxed execution AI agents can run untrusted code safely. Each component runs in an isolated sandbox with explicit permissions. True portability Deploy anywhere. No dependency hell. Same behavior locally and in the cloud. Instant deployment Tools just work. No containers, no infrastructure configuration. Publish and run.

​
Example Use Cases

​
Tools for AI Agents

Build reliable tools that your AI agents can call: API integrations, data processing, web scraping, file operations. Components provide type-safe interfaces that LLMs can understand and use correctly.

​
Polyglot Libraries

Use the right language for each job. A Rust component for performance-critical math, a Python component for ML inference, a TypeScript component for API calls—all composable in a single environment.

​
Serverless Functions

Deploy functions to the cloud without managing infrastructure. Pay only for what you use, with global edge execution for low latency.

​
Getting Started

The fastest way to get started is the Hello World guide, which walks you through creating and deploying your first component.