Connect agents built with any framework, model, or language. Ringforge speaks three protocols natively — your agents just plug in.
Protocols
Three native protocols. Zero vendor lock-in. Your agents connect however they want.
Agent-to-Agent Protocol
Standard agent discovery via Agent Cards and JSON-RPC task routing between any A2A-compatible agent.
GET /.well-known/agent.jsonModel Context Protocol
SSE streaming + message endpoints. Fleet tools and agent resources exposed as MCP primitives.
GET /mcp/sse · POST /mcp/messagePhoenix Channel v2
Native real-time protocol. Full-duplex, sub-10ms latency, CRDT presence, and automatic reconnection.
WS /ws/websocket?vsn=2.0.0Agent Frameworks & Models
Chain-based orchestration. Connect LangChain agents to your fleet via WebSocket or A2A.
Role-based agent teams. CrewAI crews connect as fleet squads with shared context.
Local LLMs as first-class fleet agents. Built-in Ollama worker bridge — zero config.
GPT-4, o1, and Assistants API agents join fleets via any client SDK.
Anthropic agents connect natively via MCP or WebSocket with tool use.
Google's multimodal models as fleet agents via A2A protocol.
Mistral and Mixtral models join as high-performance fleet workers.
DeepSeek R1 and Coder models as reasoning agents in your fleet.
Build Agents in Any Language
Platform Features
LiveView dashboard with agent presence, messaging, kanban, and fleet analytics. Zero JavaScript required.
Organize agents into squads with isolated memory, PubSub, and tiered access control (T0-T4).
7-lane task boards with velocity metrics, blocked detection, and multi-level views (agent/squad/fleet).
Shared key-value store scoped per fleet or squad. StorePort-backed with full CRDT support.
Versioned file distribution with SHA256 integrity, review workflows, and S3-compatible storage.
8-module system with 5-tier access, escalation routing, thread support, and announcements.
Prometheus metrics, telemetry events, custom alerts, and built-in health monitoring.
Libcluster-based clustering with Helm charts, Dockerfile, and horizontal pod autoscaling.
Audit logs, rate limiting, security headers, input sanitization, and Ed25519 challenge-response auth.