DecoCMS: The Context Management System

MCP-native · TypeScript-first · Deploy anywhere

Declare and Compose MCPs for your AI Agents and Workflows.

📘 Docs · 💬 Discord · 🌐 decocms.com

TL;DR:

• Centralize your company context in a private MCP mesh.
  
• Generate full-stack TypeScript AI Apps and publish them as internal MCPs.
  
• Have full control over usage, costs, monitoring and sharing.
  

🚀 What is a Context Management System?

DecoCMS is an open‑source Context Management System — the MCP Mesh for AI. It centralizes MCP connections, tools, data access, and policies to aggregate organization‑wide context in one secure place. Compose and expose virtual MCPs ("AI Apps") for any client (Claude Desktop, VS Code, custom UIs) with governance, observability, and cost control.

Think Lovable + n8n + LangGraph, running on your AWS/GCP with a single deploy command.
Full-stack AI, production-ready.

---

✨ Why DecoCMS?

AI teams are stuck between low-code prototypes and enterprise production chaos.
Backends live in n8n or LangGraph. Frontends in Lovable or React.
Deployments are separate. Auth is inconsistent. Costs spiral. Debugging is guesswork.

DecoCMS fixes that:

• 🧠 MCP-native — Compose Model Context Protocol servers with built-in policy, auth, and observability.
• ⚙️ Full-stack in TypeScript — Agents, workflows, and UIs share the same repo and types.
• 🌍 Deploy anywhere — Cloudflare Workers, AWS, or local Bun/Deno runtimes.
• 🔐 Governance built-in — RBAC, audit trails, and spend caps from day one.
• 🔭 Unified observability — Trace UI clicks → agent calls → model responses.
• 🧩 Open & modular — Install integrations, MCP tools, or full-stack modules from the Deco marketplace.

---

🧭 Architecture at a Glance

• MCP Mesh (kernel):
  • Compose and secure MCPs across your org;
  • Connect and proxy external MCPs with secure tokens;
  • Expose governed Virtual MCPs ("AI Apps") to any MCP client;
  • Enforce auth/RBAC/audit/FinOps;
  • Get full observability.
• AI App Framework (Virtual MCPs in the Mesh):
  • Build AI‑native web software (admin, workflows, custom views) that calls Tools;
  • Fullstack, From database to UI with React 19 + Tailwind v4;
  • Generative Admin interface: decopilot helps you write the PRD and implement.

---

🕸️ The MCP Mesh

The MCP Mesh is the backbone of Deco — a distributed runtime that manages context, connections, and observability for every agent in your system.

Declare and compose context — aggregate connections, authorize access, and compose tools into governed, reusable capabilities.

Core capabilities

Layer	Description
🧩 MeshContext	Unified runtime interface providing auth, storage, observability, and policy control.
⚙️ defineTool()	Declarative API for typed, auditable, observable MCP tools.
🧱 AccessControl	Fine-grained RBAC via Better Auth — OAuth 2.1 + API keys per workspace/project.
📊 OpenTelemetry	Full tracing and metrics for tools, workflows, and UI interactions.
💾 Storage Adapters	Kysely ORM → SQLite / Postgres, easily swapped.
☁️ Proxy Layer	Secure bridge to remote MCP servers with token vault + OAuth.
🧰 Virtual MCPs	Compose and expose governed toolsets as new MCP servers ("AI Apps").

On the hosted platform, usage is metered by MCP calls.

import { z } from "zod";
import { defineTool } from "~/core/define-tool";

export const CONNECTION_CREATE = defineTool({
  name: "CONNECTION_CREATE",
  description: "Create a new MCP connection",
  inputSchema: z.object({
    name: z.string(),
    connection: z.object({
      type: z.enum(["HTTP", "SSE", "WebSocket"]),
      url: z.string().url(),
      token: z.string().optional(),
    }),
  }),
  outputSchema: z.object({
    id: z.string(),
    scope: z.enum(["workspace", "project"]),
  }),
  handler: async (input, ctx) => {
    await ctx.access.check();
    const conn = await ctx.storage.connections.create({
      projectId: ctx.project?.id ?? null,
      ...input,
      createdById: ctx.auth.user!.id,
    });
    return { id: conn.id, scope: conn.projectId ? "project" : "workspace" };
  },
});

✅ Type-safe
✅ Audited
✅ Observable
✅ Callable via MCP

---

🧱 Core Architecture

(Mesh)

apps/mesh/
├── api/               # Hono HTTP + MCP proxy
├── core/              # MeshContext, AccessControl, defineTool
├── tools/             # Built-in MCP management tools
├── storage/           # Kysely DB adapters
├── auth/              # Better Auth (OAuth + API keys)
├── observability/     # OpenTelemetry tracing & metrics
└── encryption/        # Token vault & credential management

Built for scale — runs thousands of concurrent MCP connections with predictable cost and zero vendor lock-in.

---

⚙️ Developer Workflow

1. Create your project

   npm create deco
   cd my-app
   npm run dev

   → runs locally at http://localhost:8787

2. Build your stack

   • Define tools and workflows in /server
   • Add React + Tailwind UIs in /view
   • Generate typed RPC bindings with deco gen

3. Deploy anywhere

   deco deploy

   → edge-deployed via Cloudflare Workers or self-host with Bun/Deno.

---

🧩 Feature Highlights

• 🧠 Model Context Protocol (MCP) — Connect AI models to data/tools through governed context.
• 🪄 Durable Workflows — Orchestrate long-running tasks with access to any MCP.
• 🪶 Unified TypeScript Stack — One runtime for backend + frontend with typed RPC.
• 🔒 Governance & FinOps — Auth, RBAC, audit, spend caps, and policy enforcement.
• 🔭 Observability by Design — Logs, traces, and cost per step; debug entire flows visually.
• ⚡ Edge-native Deployments — Ultra-low-latency global infra via Cloudflare.
• 🧰 Deco MCP Store — Reuse full-stack modules: agents + workflows + UIs.
• 🧬 Extensible Runtime — Add your own adapters, schemas, and MCP connectors.

---

🧱 AI App Framework

Build AI‑native web software on top of the Mesh:

• React 19 + Tailwind v4 + shadcn components; design‑system powered Views.
• callTool() to invoke governed Mesh tools with types and policies applied.
• Admin, dashboards, and workflow UIs that run anywhere (edge/self‑host).

🧩 Comparison

vs Mastra

	Mastra	DecoCMS
Category	TypeScript agent/workflow framework	Full-stack MCP runtime + admin for AI apps
Focus	Build agent primitives, RAG, workflows	Ship governed agents, workflows, and UIs with RBAC, audit, FinOps
Protocol	Model/provider-agnostic (LLMs, tools)	MCP‑native runtime + proxy + policy
UI	Code-first (framework)	React/Tailwind admin shell, marketplace, settings
Observability	Tracing/evals at framework level	End‑to‑end traces UI → tools → models + spend analytics
Deployment	Node.js / serverless	Edge‑native (Cloudflare) + self‑host (Bun/Deno/AWS/GCP)

vs Metorial

	Metorial	DecoCMS
Focus	Connect AI to APIs	Build full-stack AI apps
Language	Go + TS	100 % TypeScript
Infra	Docker	Edge / Cloudflare / Self-host
Protocol	MCP clients	MCP-native runtime + admin
Auth	API keys	OAuth 2.1 + RBAC + spend caps
Observability	Dashboard	Full OpenTelemetry
UI	Separate	Shared React/Tailwind workspace
Goal	Integrate	Deploy governed AI apps fast

vs Refine

	Refine	DecoCMS
Category	React meta-framework for CRUD	Full-stack framework for AI apps
Focus	Admin panels & dashboards	Agents + Workflows + UIs
Architecture	Headless UI	Unified backend + frontend runtime
Protocol	HTTP / REST	Model Context Protocol (MCP)
Stack	React + Data Providers	TypeScript runtime + MCP Mesh
Auth	Basic auth / ACL	Better Auth (OAuth 2.1 + API keys + RBAC)
Observability	Minimal	OpenTelemetry tracing + cost analytics
Deployment	Any React env	Cloudflare edge + self-host
Use case	CRUD apps / admin dashboards	Agentic systems / governed AI apps

🧭 If you’re building internal dashboards with CRUD, use Refine. If you’re building production-scale AI apps with agents, workflows, custom UIs, and governance — use DecoCMS.

---

🧠 Tech Stack

Layer	Tech
Runtime	Cloudflare Workers / Bun / Node / Deno
Language	TypeScript (React 19 + Tailwind v4 + Zod)
Framework	Hono + Mastra + Vite
Database	Kysely → SQLite / Postgres
Auth	Better Auth (OAuth 2.1 + API keys)
Observability	OpenTelemetry + Datadog / Cloudflare Logs
Protocol	Model Context Protocol (MCP)

---

🪪 License & Partners

DecoCMS ships with a Sustainable Use License (SUL). Read the full terms in LICENSE.md.

For Partners (Service Providers, Software Houses, SIs, Digital Agencies)

• Free to self‑host and use to deliver client projects, as long as each deployment is for the client’s internal use (see SUL §3.3).
• Build, implement, and charge for your services — go make money, no questions asked.
• Keep your workspace private — bring your own models and data.

Do you implement Agentic Software? We have customers for you. Email us.

For Enterprises (Mission‑Critical or Revenue‑Generating)

• These cases require a commercial self‑hosted Enterprise license per the SUL.
• We help you deploy at scale in your cloud (AWS or GCP), with governance, observability, and performance tuned for large workloads.
• Get production support and reliability.

Questions? Contact us at [email protected] or visit decocms.com.

---

🤝 Contributing

We welcome contributions from vibecoders, agentic engineers, and builders of the next internet. See CONTRIBUTING.md for setup, coding standards, and roadmap.

🧭 Upcoming milestones

• Multi-tenant admin dashboard
• MCP store
• Edge debugger / live tracing
• Native view components as tools

---

_{Made with ❤️ by the deco community.
Building the open-source operating system for AI-native apps — secure, scalable, and governed by context.}