Gradio
Build interactive machine learning interfaces with Python
agentOS
Embed an isolated operating environment for AI agents in your backend
Meet agentOS
agentOS is a library from Rivet that gives an AI agent files, processes, shell tools, and networking inside an isolated runtime. Developers embed it in a backend process and connect it through ordinary function calls. The runtime combines WebAssembly with V8 isolates to support workloads without assigning a separate virtual machine to every agent.
An agent environment can run supported Node.js, Python, and Bash tools, use subprocesses, and access storage mounted from services such as S3 or Google Drive. Trusted host functions expose specific application operations without handing raw credentials to the agent. Network access and other capabilities require explicit configuration. Compatibility and isolation boundaries depend on the runtime, packages, and permissions selected for the workload.
Discussion
Tried it? Share a useful thought or ask a question.
Good conversations start with a hello.
Sign in to commentProduct details
- Website
- rivet.dev
- Status
- Launched
- Category
- Developer tools
- Listed
Explore related topics
Similar products
Django
Build Python web applications with integrated data models, templates and administration tools.
Zerolang
Explore a programming language built around checked graph edits
Thesys
Build AI interfaces that answer with interactive components and visual outputs.
AUDR
An open record format for attributing AI-agent usage and costs
Lloyal
Build applications with local models and agents inside the runtime.