Atmosphere is a Java framework for building streaming AI agents on the JVM. It lets developers declare agent behavior with an @Agent annotation while the framework handles transport, streaming, tool calls, memory, reconnect behavior, authorization, and observability. A single agent can be exposed over WebSocket, Server-Sent Events, long polling, gRPC, and WebTransport over HTTP/3 depending on the modules included. It also supports agent-facing protocols such as MCP, A2A, and AG-UI, along with external messaging channels such as Slack, Telegram, Discord, WhatsApp, and Messenger. The project is built for teams that need AI agents to behave like production services instead of simple chat demos. It is useful for JVM developers who want portable, observable, real-time AI agent endpoints across multiple clients and protocols.

Features

  • JVM streaming agent framework
  • WebSocket, SSE, gRPC, long-polling, and HTTP/3 transports
  • Annotation-based agent declaration
  • MCP, A2A, and AG-UI protocol support
  • Memory, tool calling, and reconnect handling
  • Observability, authorization, and cost tracking

Project Samples

Atmosphere Screenshot 1

Project Activity

See All Activity >

Categories

Agentic AI

License

Apache License V2.0

Follow Atmosphere

Atmosphere Web Site

Other Useful Business Software
Train ML Models With SQL You Already Know Icon
Train ML Models With SQL You Already Know

BigQuery automates data prep, analysis, and predictions with built-in AI assistance.

Build and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
Start Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of Atmosphere!

Additional Project Details

Programming Language

Java

Related Categories

Java Agentic AI Tool

Registered

2026-05-18
Report inappropriate content