Agent-Native

Open-source framework for building agentic apps

Open-source TypeScript framework for building AI agents with rich UIs, one action powering every surface.

Agent-Native homepage

Agent-Native homepage

1 / 2

Overview

What is Agent-Native?

Open-source framework for building agentic apps

Agent-Native is an open-source TypeScript framework for building agentic apps — AI agents with rich, intuitive UIs. Define an action once with defineAction() and reuse the same code across the agent, React UI, HTTP API, MCP server, A2A endpoint and CLI. Agents share application state and data with the UI, so anything the UI can do, the agent can do. It ships with built-in authentication, organizations and permissions, skills and memory so agents learn how your team works, agent teams over the A2A protocol, and event-driven automation. Model-agnostic: bring your own LLM, build locally with npx, deploy anywhere. MIT licensed.

Platforms and languages

Agent-Native Availability

Platforms

WebApi

Languages

EnglishSpanishFrenchGermanPortugueseChineseJapaneseKoreanHindiArabic

Capabilities

Agent-Native Key Features

One action powers every surface

Define a capability once with defineAction() and reuse the same code across the agent, React UI, HTTP API, MCP, A2A and CLI.

Agent chat embedded in React UI

Agents share application state and data with the UI — anything the UI can do, the agent can do.

Auth, orgs and permissions

Built-in authentication, organizations and permissions with sharing and real-time collaboration.

Skills and memory

Agents learn how your team works and remember context across sessions.

Agent teams and automation

Coordinate multiple agents over the A2A protocol with event-driven automation workflows.

Best for

Who uses Agent-Native?

Developers

building agentic apps with shared actions across agents, UI and APIs

Plans and access

Agent-Native Pricing

Free

Free and open source under the MIT License

Common questions

Agent-Native FAQs

Is Agent-Native free?

Yes. Agent-Native is open source under the MIT License; install it locally with npx and deploy anywhere.

How does one action power every surface?

Define a capability once with defineAction(), then reuse the same code across the agent, React UI, HTTP API, MCP server, A2A endpoint and CLI — one implementation powers every surface.

Which AI models can I use?

Any model. Agent-Native is model-agnostic — bring your own LLM and deploy anywhere.

What comes built in?

The framework wires UI, context, data, permissions and infrastructure together, with built-in authentication, organizations, permissions, real-time collaboration, skills and memory.

Reviews

See what the community thinks and share your experience.

—

Based on 0 ratings

Rating distribution

0
0
0
0
0

Leave a review

Sign in to rate

Community reviews

No reviews

No written reviews yet.

Explore by category

AI Development Infrastructure

View all AI Development Infrastructure websites
Linkdash preview
Linkdash

Short links your AI agents can drive. Turn a URL, file, or text snippet into a tracked link, with A/B variants, targeting, and live analytics, from a dashboard,

AI Development Infrastructure
GPUBank preview
GPUBank

GPUBank is a cloud GPU rental platform offering affordable, on-demand NVIDIA and AMD GPUs for AI training, machine learning, LLM inference, and generative AI wo

AI Development Infrastructure

Explore by category

AI Apis Sdks

View all AI Apis Sdks websites
LurkAPI preview
LurkAPI

A social media data API and MCP server: posts, ads, profiles and transcripts from 16 platforms.

AI Apis Sdks
Lava MCP preview
Lava MCP

News-flow heat data for China A-share stocks and concept sectors via REST API, remote MCP server and Agent Skill, for quant research and AI agents.

AI Apis Sdks