dannote

dannote / live_render

Public

Server-driven generative UI for Phoenix LiveView

37
2
100% credibility
Found Mar 09, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Elixir
AI Summary

LiveRender enables AI-generated user interfaces in Phoenix LiveView apps by rendering JSON specifications as server-side function components with built-in safety guardrails.

How It Works

1
🔍 Discover LiveRender

You hear about a cool way to let AI design beautiful screens for your web app, making it easy to create dynamic dashboards or chats without drawing every layout yourself.

2
📦 Add it to your app

You simply include this helper in your existing web project, and everything is ready to start building smarter interfaces.

3
🎨 Pick your building blocks

You choose safe, ready-made pieces like cards, buttons, and charts that the AI can mix and match to build your screens.

4
🤖 Connect your AI helper

You link up your favorite AI service so it understands exactly what pieces it can use and how to arrange them.

5
💭 Describe what you want

You type a simple idea like 'create a weather dashboard with metrics and trends', and the AI starts generating the layout right before your eyes.

6
✨ Watch it come alive

The screen updates smoothly as the AI builds cards, buttons, and data displays, freezing stable parts while new bits appear progressively.

🚀 Enjoy AI-powered magic

Your app now has lively, intelligent interfaces that adapt to prompts, saving you hours of design work and delighting your users.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 37 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is live_render?

LiveRender brings server-driven generative UI to Phoenix LiveView in Elixir, letting AI models spit out JSON specs that render directly into functional components server-side, with only HTML diffs pushed over WebSocket. It solves the mess of client-side JSON parsing by keeping everything on the BEAM—no runtime bloat for the browser—while supporting live rendering from LLMs like Claude. Think Vercel json-render, but idiomatic for Phoenix: register components, feed a prompt, and watch dynamic UIs like dashboards or chats emerge progressively.

Why is it gaining traction?

It stands out with guardrails—AI sticks to your component catalog, avoiding hallucinated junk—plus streaming patch mode for buttery-smooth progressive builds as tokens arrive. Built-in shadcn-compatible components (grids, metrics, tabs) and easy LLM hooks (ReqLLM, Jido) mean you get server-driven UI like React or Android libs, but with LiveView's real-time magic and data binding. Devs dig the safety and zero-config start: prompt → spec → render.

Who should use this?

Phoenix LiveView devs building AI-driven apps like agentic chats, dynamic dashboards, or live rendering previews—think weather tools or crypto trackers where LLMs generate layouts on-the-fly. Ideal for backend teams wanting generative UIs without frontend JSON parsers, or experimenters porting server-driven UI from React GitHub repos to Elixir.

Verdict

Promising for early adopters, with solid docs, example app, and MIT license, but 19 stars and 1.0% credibility scream "alpha"—test thoroughly before prod. Grab it if you're prototyping live rendering AI in Phoenix; skip for mission-critical unless you contribute.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.