About Flezi AgentBox
An AI Agent Builder and Marketplace platform that empowers developers and creators to build, share, and run intelligent agents.
Our Mission
We believe AI agents should be accessible to everyone. Flezi AgentBox provides a complete platform for the AI agent ecosystem — from a visual builder and marketplace, to a Go CLI with Docker-style UX, to kernel-isolated NemoClaw sandbox execution. With Agent Autopilot (v3.0), your agents run autonomously in the background — scanning code, reviewing PRs, blocking secrets, and generating reports. Whether you're a solo developer or managing an enterprise team, Flezi AgentBox is your AI-powered dev team.
By The Numbers
478
Features Shipped
187
Epics Completed
4,394+
Tests Passing
54/54
OWASP Score
Technology Stack
Built with modern, battle-tested technologies for performance, security, and developer experience.
Team
Admin
Project Lead & Full-Stack Developer
Architect of the Flezi AgentBox platform. Designed the system from PRD to production across 187 epics using the BMAD methodology — from marketplace MVP to autonomous agent autopilot with multi-agent pipelines.
Project Milestones
v1.0
Core Platform
Epics 1-129
Auth, marketplace, builder lab, Docker sandbox, multi-LLM, BYOK vault, E2E tests, OWASP hardening, production deployment
v1.1
NemoClaw Mandatory
Epics 130-135
Kernel-level sandbox — Landlock + seccomp + netns, L0-L3 permissions, binary integrity, IPC protocol
v1.5
CLI Production
Epics 136-154
Go CLI, agent package format, MCP tool-calling, E2E demo, Windows support, rate limit retry
v1.6
MCP Local Server
Epics 155-157
Universal MCP server (stdio + HTTP), agent-as-tool, hot reload, agent mesh, LAN discovery
v1.7-1.9
Intelligence
Epics 158-164
Context-aware execution, CLI UX redesign (agentbox <agent>), 3-layer agent memory, maturity L1-L5
v2.0-2.1
Autonomy
Epics 165-172
Autonomous workflows, pre-commit scanning, agent factory, self-update with A/B testing
v2.4
Experience
Epics 173-176
Web UI, interactive playground, before/after showcase, onboarding wizard, admin dashboard
v3.0
Agent Autopilot
Epics 177-187
Background daemon, event router, execution queue, multi-agent pipelines, PR review bot, budget control, approval workflows, mission control dashboard
Contact
Have questions or want to get in touch? Use the feedback widget (bottom-right corner) to send us a message, or visit our GitHub repository for source code and issue tracking.