Atmosphere Real-Time JVM Agent Engine

What does this agent do, and when should you use it?

Atmosphere is a real-time, event-driven JVM framework for running AI agents as application services. Developers declare an agent with @Agent and add classpath modules to expose browser endpoints, MCP, A2A, AG-UI, or messaging channels. Its atmosphere-ai module uses the AgentRuntime SPI to connect a built-in OpenAI-compatible runtime and multiple external runtime adapters, while retaining the native framework's model, tool, and memory facilities. Tokens and events move through a broadcaster pipeline over WebSocket, SSE, long-polling, or gRPC; WebTransport over HTTP/3 is optional and has additional classpath and development-certificate requirements. The repository includes durable sessions, checkpoints, human approval, policy controls, auditing, and multi-agent coordination, but does not host model weights, own compute scheduling, or provide payments and commerce primitives. It fits teams with an existing Java, Spring Boot, Quarkus, or Servlet-container stack that need governed, long-lived agent interactions.

An application marks a class containing @Prompt, @Command, and @AiTool methods with @Agent. atmosphere-agent registers /atmosphere/agent/my-agent for browser access, streaming AI dispatch, memory, commands, and /help; adding atmosphere-mcp, atmosphere-a2a, or atmosphere-agui registers their respective protocol endpoints. On a request, AgentRuntime selects an available runtime adapter to call a model; tool calls can pass through ToolExecutionHelper.executeWithApproval, while policy and @AgentScope can reject requests before runtime dispatch. Sessions can persist through SQLite or Redis, and checkpointed workflows can hibernate at a human-approval gate and resume through REST. The broadcaster pipeline delivers text and run events to connected clients.

1. A Spring Boot team building a customer-support agent that streams tool-assisted responses to a web client over WebSocket or SSE.
2. A Java developer building a startup-team or specialist collaboration flow with @Coordinator, AgentFleet, handoffs, and durable checkpoints.
3. An enterprise team that must apply approval, scope restrictions, PII rewriting, or cost ceilings before email, database, or other sensitive tool calls.
4. An integration team that needs one Java agent exposed simultaneously to browser clients, MCP clients, A2A callers, and Slack, Telegram, Discord, WhatsApp, or Messenger.
5. A workflow developer whose long-running interactions must survive disconnects and later be replayed, approved, or resumed through REST.

How do you install or deploy this agent?

Install the CLI:

brew install Atmosphere/tap/atmosphere

Or:

curl -fsSL https://raw.githubusercontent.com/Atmosphere/atmosphere/main/cli/install.sh | sh

Create and run an app:

atmosphere new my-agent --template ai-chat
cd my-agent
LLM_API_KEY=your-key ./mvnw spring-boot:run

The documented baseline is Java 21+, with Spring Boot 4.0.7 or Spring Boot 3.5 using -Pspring-boot3, and Quarkus 3.36.0+.

How do you use this agent?

For a sample, run atmosphere run spring-boot-multi-agent-startup-team. In an application, add atmosphere-spring-boot-starter and atmosphere-agent, then annotate a class with @Agent(name = "my-agent", description = "What this agent does"); use @Prompt for messages, @Command for commands, and @AiTool for tools. The default browser endpoint is /atmosphere/agent/my-agent; adding atmosphere-mcp exposes /atmosphere/agent/my-agent/mcp. Scaffold against Spring AI with --runtime spring-ai, or replace a sample's pinned runtime with --runtime langchain4j --force.

What are this agent's strengths and limitations?

How does this agent compare with similar options?

Key facts side by side with the most closely related agents.

How does FollowAgents rate this agent?

Recent releases

Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.

1. atmosphere-4.0.76 Atmosphere 4.0.76 2026-10-22
   Release notes excerpt

   Added - keep a prompt under the run id its transport mints, shown again on reload at once - send a prompt under a run id the client mints, kept before it is sent - adopt the run id a client mints for its prompt, running a prompt once per id - serve the Atmosphere Console as the sample UI - keep the picked room across a page reload Fixed - ai-tools lets atmosphere.ai.mode and model apply when LLM_MODE and LLM_MODEL are unset - log a refused cleartext key once and a tool failure without a stack - new --runtime without --force warns when the template pins another adapter …

   View on GitHub ↗
2. atmosphere-4.0.75 Atmosphere 4.0.75 2026-10-17
   Release notes excerpt

   Added - every member of a broadcastReply room sees the prompts the others type - a page reloaded mid-answer reattaches to its AI run - name the timeout cancel reason in the client type and the Console - add a cancellable model call to the reattach harness - produce the application's AiEnvironment as a CDI bean Fixed - a Clear before the first token discards this client's room turn, not another member's - a room prompt is echoed only from the admission of the turn that streams it - a reattached run streams like a turn the page started …

   View on GitHub ↗
3. atmosphere-4.0.74 Atmosphere 4.0.74 2026-10-16
   Release notes excerpt

   Added - deprecate the static AI readers and binders; samples read their own AiEnvironment - give each application its own native clients in every runtime adapter - give each application its own AI knobs, settings and runtime instances - enforce both an explicit skill scopeTier and a restrictive @AgentScope, stricter wins - let one ScopePolicy enforce several scope legs that a request must all pass - mark the pagehide close beacon going-away, keep close() a leave - give a page reload over SSE, long-polling or streaming the reattach grace …

   View on GitHub ↗

FAQ