Agent Studio

Visual AI agent builder with multi-agent orchestration and continuous learning.

---

Quick Start

git clone https://github.com/webdevcom01-cell/agent-studio.git
cd agent-studio
cp .env.example .env
# Fill in DEEPSEEK_API_KEY and OPENAI_API_KEY, then:
docker compose up

Open http://localhost:3000 and create your first agent.

No Docker? See Manual Setup below.

---

Features

• Visual Flow Editor — Drag-and-drop builder with 32 node types (AI, logic, integrations, webhooks) powered by XyFlow
• Enterprise RAG Pipeline — Ingest URLs, PDFs, DOCX; chunk with 5 strategies; hybrid search (semantic + BM25) with pgvector; LLM re-ranking
• MCP + A2A Protocols — Connect external tools via Model Context Protocol; agent-to-agent communication following Google A2A v0.3
• ECC Developer Skills — 60+ skill modules and 25 developer agent templates with autonomous meta-orchestration and continuous learning
• CLI Generator — 6-phase AI pipeline wraps any CLI as an MCP server (Python FastMCP or TypeScript MCP SDK)
• Agent Evals — 3-layer testing: deterministic assertions, semantic similarity, LLM-as-Judge with 12 assertion types and deploy-triggered runs
• Agent Marketplace — 137 templates across 12 categories with faceted search, discovery, and one-click import
• Embeddable Chat Widget — Drop-in widget for any website with streaming responses, customizable appearance, and mobile support

---

Architecture

graph TB
    subgraph Client
        UI[Next.js App Router]
        FE[Flow Editor - XyFlow]
        Chat[Chat Interface]
    end

    subgraph API["API Layer (50+ routes)"]
        Agents[Agent CRUD]
        FlowAPI[Flow Versioning & Deploy]
        ChatAPI[Chat - Streaming NDJSON]
        KBAPI[Knowledge Base]
        Evals[Eval Runner]
        Webhooks[Inbound Webhooks]
        CLI[CLI Generator]
    end

    subgraph Runtime["Flow Runtime Engine"]
        Engine[Execution Loop]
        Handlers[32 Node Handlers]
        Stream[Streaming Engine]
    end

    subgraph AI["AI Layer"]
        SDK[Vercel AI SDK v6]
        Providers[7 Providers / 18 Models]
        MCP[MCP Client + Pool]
        AgentTools[Agent-as-Tool]
    end

    subgraph Data
        PG[(PostgreSQL + pgvector)]
        Redis[(Redis)]
        KB[RAG Pipeline]
    end

    subgraph ECC["ECC Module"]
        Skills[60+ Skills]
        Meta[Meta-Orchestrator]
        Learn[Instinct Engine]
        ECCMCP[Skills MCP Server]
    end

    UI --> API
    FE --> FlowAPI
    Chat --> ChatAPI
    API --> Runtime
    Runtime --> AI
    AI --> Providers
    AI --> MCP
    Runtime --> Data
    KB --> PG
    ECC --> ECCMCP
    Meta --> AgentTools

Loading

---

Manual Setup

Setup without Docker

Prerequisites

• Node.js 20+
• pnpm 9+
• PostgreSQL with pgvector extension

Steps

# Install dependencies
pnpm install

# Configure environment
cp .env.example .env.local
# Required: DATABASE_URL, DIRECT_URL, DEEPSEEK_API_KEY, OPENAI_API_KEY,
#           AUTH_SECRET, AUTH_GITHUB_ID/SECRET or AUTH_GOOGLE_ID/SECRET

# Enable pgvector (run in your PostgreSQL client)
# CREATE EXTENSION IF NOT EXISTS vector;

# Setup database and generate client
pnpm db:push && pnpm db:generate

# Start dev server
pnpm dev

Open http://localhost:3000.

---

Available Commands

pnpm dev              # Dev server (Turbopack)
pnpm build            # Production build
pnpm lint             # ESLint
pnpm typecheck        # TypeScript check
pnpm test             # Vitest unit tests (1500+)
pnpm test:e2e         # Playwright E2E tests
pnpm db:push          # Sync Prisma schema to DB
pnpm db:generate      # Generate Prisma client
pnpm db:studio        # Prisma Studio UI

---

Tech Stack

Layer	Technology
Framework	Next.js 15.5, App Router, Turbopack
Runtime	React 19
Language	TypeScript strict
Styling	Tailwind CSS v4
Database	PostgreSQL + pgvector, Prisma v6
AI	Vercel AI SDK v6 (7 providers, 18 models)
Auth	NextAuth v5 (GitHub + Google OAuth)
Flow Editor	@xyflow/react v12
MCP	@ai-sdk/mcp (Streamable HTTP + SSE)
Validation	Zod v3
UI	Radix UI + lucide-react
Tests	Vitest (unit) + Playwright (E2E)

---

Project Structure

prisma/schema.prisma        # 30+ models, pgvector, versioning, A2A, ECC
src/
  app/                      # Pages and 50+ API routes
    builder/[agentId]/      # Flow editor
    chat/[agentId]/         # Chat interface
    knowledge/[agentId]/    # Knowledge base
    evals/[agentId]/        # Agent evals
    discover/               # Agent marketplace
    skills/                 # ECC Skills Browser
    cli-generator/          # CLI-to-MCP pipeline
  components/               # React components
  lib/
    runtime/                # Flow engine (32 handlers)
    knowledge/              # RAG pipeline
    ecc/                    # ECC module
    evals/                  # Eval runner
    mcp/                    # MCP client + pool
  data/                     # Agent templates
services/ecc-skills-mcp/    # Python FastMCP server
e2e/                        # Playwright E2E tests
docs/                       # Documentation

---

Documentation

Document	Description
Platform Overview	Features and architecture
Getting Started	Setup guide
Node Reference	All 32 node types
Knowledge Base Guide	RAG pipeline
CLI Generator	MCP bridge generation
Agent Evals	Testing framework
CHANGELOG	Version history

---

Contributing

Contributions are welcome. See CONTRIBUTING.md for guidelines.

---

License

Apache License 2.0

Copyright 2026 Agent Studio Contributors