Overview
High-level architecture overview of Archestra Platform components
Archestra is a centralized AI platform for organizations where engineers and non-technical teams both work with AI agents. A non-technical user works through a chat UI and gets results right away. An engineer builds agents in LangChain, n8n, Python, or another stack, using the MCP orchestrator, guardrails, and observability. Both use the same platform.
Fun fact: the team behind Archestra.AI previously worked on Grafana OnCall.
Composable Components
Archestra is a set of composable components. Most organizations already run tools like n8n, LiteLLM, Grafana, or custom MCP servers. Adopt all of Archestra, a few components, or just one — each works with what you already have.
Agentic Chat — ChatGPT-like interface for non-technical users. Talk to agents via web UI, Slack, MS Teams, or Email.
Agent Runtime — No-code builder for autonomous agents. Define system prompts, assign MCP tools and sub-agents, configure triggers.
MCP Orchestrator — Run MCP servers as isolated pods in Kubernetes.
Knowledge Base — Built-in RAG Knowledge Base to give your agents access to your data.
LLM & MCP Proxies — Drop-in proxy between your apps and LLM providers. MCP Gateway provides a single endpoint for all MCP tools. Works with any framework: n8n, LangChain, Vercel AI, Pydantic AI, Mastra.
Security & Guardrails and Observability — Deterministic tool invocation policies and trusted data policies that cannot be bypassed by prompt injection. Prometheus metrics, OpenTelemetry tracing, and per-team cost tracking.
See Pricing Model for licensing details.