Nexent

Active
GitHub Python MIT

Description

A zero-code platform for auto-generating production-grade AI agents using Harness Engineering principles with unified tools, skills, memory, and orchestration with built-in constraints and feedback loops.

Key Features

  • Zero-code platform for auto-generating production-grade AI agents from natural language
  • A2A (Agent-to-Agent) protocol for seamless multi-agent cooperation
  • Layered memory mechanism with user-level and user-agent-level persistence
  • Progressive Skill disclosure maximizing context window efficiency
  • Personal-grade knowledge base supporting 20+ document formats with fine-grained access control
  • Full multimodal interaction: voice, text, images, and files

Use Cases

💡 Rapid agent creation without coding for business users
💡 Multi-agent collaborative workflows with distributed task execution
💡 Enterprise knowledge management with versioned agent marketplace
💡 Voice-enabled AI assistants with multimodal capabilities
💡 Production deployment on Docker or Kubernetes for teams of any size

Tags

zero-code agent-platform harness-engineering orchestration python

Categories

⚡ Agent Tools

Quick Start

Docker: git clone https://github.com/ModelEngine-Group/nexent.git && cd nexent/docker && bash deploy.sh. K8s: cd nexent/k8s/helm && ./deploy.sh. Or try the online demo at nexent.tech. Minimum: 4 CPU cores, 8 GiB RAM, 40 GiB disk.
Visit GitHub Visit Website

Related Projects