Everything you need to build, orchestrate, and deploy multi-agent AI systems. From protocols to production infrastructure.
Layer 1
The brain that coordinates all AI agents
Multi-agent orchestration platform with persistent memory, session management, and peer mesh networking.
Layer 2
Standards for agent communication and identity
Agent Messaging Protocol — secure, cryptographically signed agent-to-agent messaging with Ed25519 signatures.
Authentication protocol for AI agents with lifecycle management, OIDC token exchange, and audit trails.
Layer 3
Build specialized AI agents and assistants
Layer 4
Connect AI agents to communication platforms
Bridge AI agents to Discord servers with prompt injection defense and multi-agent routing.
Connect AI agents to Slack workspaces with secure message routing.
Connect AI agents to WhatsApp Business with bidirectional messaging.
Layer 5
Production-ready APIs your agents use to build
Fully-managed API blocks — Auth, Files, CRM, Forms, Content, Sales, Rewards, AI, and more. 203 agent-ready skills.
Install any block as a Claude Code MCP server. Each block is a standalone plugin with typed tools.
The ecosystem layers work together to power multi-agent applications
Claude Code, Codex, Aider, Cursor, or any terminal-based agent
Routes messages, manages sessions, coordinates multi-agent workflows
Secure messaging
Discord, Slack, Email, WhatsApp
Auth, Files, CRM, Forms, Content, Sales, Rewards, AI, and more
All open source, all MIT licensed. Pick the pieces you need and start shipping.
Get product updates, engineering posts, and new block announcements delivered to your inbox.