Flezi AgentBox

Platform Overview

Learn what Flezi AgentBox is, its core capabilities, and how it can help you build and deploy AI agents

Platform Overview

Welcome to Flezi AgentBox — an AI Agent Builder and Marketplace platform that enables you to create, publish, discover, and execute intelligent agents.

What is Flezi AgentBox?

Flezi AgentBox is a comprehensive platform for the AI agent ecosystem. Whether you're a developer building custom agents, a business looking for ready-made AI solutions, or a creator monetizing your AI expertise, Flezi AgentBox provides the tools you need.

Core Capabilities

Agent Builder Lab

The Builder Lab is a three-phase guided workflow for creating AI agents:

  1. Discovery Phase — Define your agent's purpose, capabilities, and knowledge base
  2. Logic Phase — Design the processing pipeline with a visual DAG editor
  3. Synthesis Phase — Generate the final agent blueprint and publish

New to building agents? Check out the Quickstart Guide to create your first agent in 5 minutes.

Marketplace

Browse, discover, and acquire AI agents created by the community:

  • Search & Filter — Find agents by category, runtime, access level, and more
  • Reviews & Ratings — Community-driven quality signals
  • Creator Storefronts — Follow your favorite creators
  • Agent Collections — Curate and share themed agent bundles

Agent Pipelines

Chain multiple agents together for complex workflows:

yaml
pipeline:
  name: "Code Review Pipeline"
  stages:
    - agent: "code-analyzer"
      parallel: false
    - agent: "security-scanner"
      parallel: true
    - agent: "report-generator"
      parallel: false

Execution Sandbox

Run agents in a secure, isolated environment with:

  • Real-time streaming output via SSE
  • Keyboard shortcuts for power users
  • Execution history and environment profiles
  • Docker-based container isolation

MCP IDE Integration

Invoke Flezi AgentBox agents directly from your IDE (VS Code, Cursor, Claude Code) via the Model Context Protocol server.

Architecture Overview

Flezi AgentBox follows a modern microservices architecture:

  • Frontend — Next.js 16 with server components and Shadcn/UI
  • Orchestrator — Fastify 5 with Docker container management
  • Database — Supabase (PostgreSQL) with Row Level Security
  • Deployment — Docker Compose with health checks

Next Steps

Create your first agent

Head to the Quickstart Guide for a step-by-step tutorial.

Explore the marketplace

Browse available agents in the Marketplace Guide.

Learn core concepts

Understand agents, pipelines, and templates in the Core Concepts page.