Skip to main content
Agent Stack is open, self-hostable infrastructure for deploying AI agents built with any framework. Hosted by the Linux Foundation and built on the Agent2Agent Protocol (A2A), it gives you everything needed to move agents from local development to shared production environments—without vendor lock-in.

​
What Agent Stack Provides

Everything you need to deploy and operate agents in production:
  • Self-hostable server to run your agents
  • Web UI for testing and sharing deployed agents
  • CLI for deploying and managing agents
  • Runtime services your agents can access:
    • LLM Service — Switch between 15+ providers (Anthropic, OpenAI, watsonx.ai, Ollama) without code changes
    • Embeddings & vector search for RAG and semantic search
    • File storage — S3-compatible uploads/downloads
    • Document text extraction via Docling
    • External integrations via MCP protocol (APIs, Slack, Google Drive, etc.) with OAuth
    • Secrets management for API keys and credentials
  • SDK (agentstack-sdk) for standardized A2A service requests
  • HELM chart for Kubernetes deployments with customizable storage, databases, and auth
Build your agent using LangGraph, CrewAI, or your own framework—SDK handles runtime service requests automatically.

​
How It Works

  1. Build your agent using agentstack-sdk
  2. Deploy with a single CLI command
  3. Users interact through the auto-generated web UI
Your agents request infrastructure services at runtime through A2A protocol extensions.
  • Development: Run locally with full services for rapid iteration
  • Production: Deploy to Kubernetes via HELM and integrate with your infrastructure

​
Get Started

Quickstart

Get up and running in one command

Connect A2A Agents

Run existing A2A-compatible agents for instance GUI

Start Building Agents

Build your first agent with the SDK

Deploy to Production

Deploy Agent Stack to Kubernetes for your team