ANA — Agent‑Native Agent

Build apps you operate by watching and talking.

English · 한국어

TL;DR

ANA is an Agent‑Native Agent for Agent‑Native Lifestyle (ANL). It is a self‑hosted app you operate by watching a live dashboard and conversing with a coding agent that serves as the runtime. Need a new behavior? Ask once — ANA proposes the change, applies it on approval, and evolves the app at runtime. No PR, no ship step — the running agent rewrites the app live and the dashboard reloads.

Use = Build. That's the whole idea.

Why Agent, Not Assistant

An assistant waits for instructions. An agent uses judgment, executes, and improves the tools around your work. Every tool today still forces a trade‑off between using and building — ANA collapses that gap.

SaaS is instant but frozen. Coding agents are infinitely malleable but build‑time only — you ship, then use. ANA makes using the app (talking) the same act as building it (changing behavior), because the agent is native to the runtime.

The Three Principles

1. Watch + Converse — visual state and chat live in one view. You operate by looking and talking, not clicking through fixed UI.
2. Agent as Runtime — the agent reads your data, acts, and rewrites the app's own code when asked. Inference is the runtime.
3. Own Your Harness — zero dependencies, self‑hosted, yours forever. It keeps evolving with you.

These are also the acceptance criteria for every ANA built with this harness.

How it works

flowchart TB
  subgraph UX["Watch + Converse"]
    direction LR
    U["User"] <--> D["Dashboard"]
  end

  subgraph RT["Agent Runtime"]
    direction LR
    B["Bridge"] --> C["Channel"]
    C --> A["Coding Agent"]
    A --> B
  end

  D -->|"intent"| B
  B -->|"proposal + sync"| D

Inbound messages travel through the channel. Outbound agent responses return through the dashboard API, so ANA can show rich before/after proposals and approval cards. State is versioned, and every device syncs. The agent is the backend — there's no separate server logic to write; you grow it by talking.

Install the skills (30 s)

Prerequisite: Claude Code installed. Want a running app first, no setup? Jump to Start from a template → node server.js.

git clone https://github.com/tykimos/agent-native-agent
cp -r agent-native-agent/skills/* ~/.claude/skills/

Then, in Claude Code, just describe the app:

"Build a weekly family planner as an agent native agent"
"Add voice input to this ANA"        # ← evolve: one sentence, no deploy
"Put an at-a-glance progress bar on top"

The agent-native-app-harness orchestrator skill triggers and builds your ANA: define one screen → design → wire up → run the evolution loop. Step‑by‑step in build-workflow.md.

Start from a template

Don't want to build from an empty screen? ana‑starter is a ready‑to‑run ANA — the same design system, logo, and runtime as the reference dashboard — with a simple menu you grow by talking.

# GitHub → "Use this template", then in your clone:
node server.js        # → http://localhost:8777   (npm start / node start.js also work)

It ships the dashboard + fakechat relay. Connect a Claude Code session with the fakechat channel to complete the watch→converse loop (npm run all starts server + relay together). Details in the starter README.

Building blocks

The demo GIF at the top is the "Work Secretary" ANA — six channels (mail · Slack · KakaoTalk · approvals · calendar · SMS) collapsed into one board, sorted by urgency, evolved by talking.

ANA and ANL

ANA enables ANL. Real ANL cases — lifestyles built with ANA — live in the companion repo: agent‑native‑lifestyle.

Star History

Contributing

ANA is meant to be owned and evolved — that includes this repo. Issues, ideas, and PRs are welcome; see CONTRIBUTING.md for how to add building blocks or examples.

If you build something with ANA, add it to the agent‑native‑lifestyle gallery so ANL stays visible as real usage.

If ANA changes how you think about apps, ⭐ star the repo so others can find it.

License

MIT © tykimos