{"id":51458881,"url":"https://github.com/narrowin/awesome-generative-ui","last_synced_at":"2026-07-06T03:30:39.381Z","repository":{"id":367332669,"uuid":"1117387225","full_name":"narrowin/awesome-generative-ui","owner":"narrowin","description":"A curated list of resources for AI-generated user interfaces — systems where LLMs dynamically create, compose, and render UI components.","archived":false,"fork":false,"pushed_at":"2026-06-25T13:48:18.000Z","size":38,"stargazers_count":52,"open_issues_count":2,"forks_count":5,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-25T15:32:45.113Z","etag":null,"topics":["awesome","awesome-list","experiment","frontend-development","generative-ai","research","ui","ui-design"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"cc-by-4.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/narrowin.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-16T08:42:52.000Z","updated_at":"2026-06-25T13:48:55.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/narrowin/awesome-generative-ui","commit_stats":null,"previous_names":["narrowin/awesome-generative-ui"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/narrowin/awesome-generative-ui","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/narrowin%2Fawesome-generative-ui","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/narrowin%2Fawesome-generative-ui/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/narrowin%2Fawesome-generative-ui/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/narrowin%2Fawesome-generative-ui/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/narrowin","download_url":"https://codeload.github.com/narrowin/awesome-generative-ui/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/narrowin%2Fawesome-generative-ui/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35177220,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-06T02:00:07.184Z","response_time":106,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["awesome","awesome-list","experiment","frontend-development","generative-ai","research","ui","ui-design"],"created_at":"2026-07-06T03:30:38.776Z","updated_at":"2026-07-06T03:30:39.368Z","avatar_url":"https://github.com/narrowin.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Awesome Generative UI [![Awesome](https://awesome.re/badge.svg)](https://awesome.re)\n\n\u003e A curated list of resources for AI-generated user interfaces — systems where LLMs dynamically create, compose, and render UI components.\n\nGenerative UI represents a paradigm shift from \"AI assists coding\" to \"AI generates interfaces directly.\" Instead of writing components, you describe what you need and the AI produces a working UI.\n\n## Contents\n\n- [What is Generative UI](#what-is-generative-ui)\n- [Key Concepts](#key-concepts)\n- [Research](#research)\n- [Benchmarks \u0026 Datasets](#benchmarks--datasets)\n- [Specifications \u0026 Protocols](#specifications--protocols)\n- [Frameworks \u0026 SDKs](#frameworks--sdks)\n- [Tools \u0026 Platforms](#tools--platforms)\n- [Security \u0026 Sandboxing](#security--sandboxing)\n- [Evaluation \u0026 Testing](#evaluation--testing)\n- [Schemas \u0026 Patterns](#schemas--patterns)\n- [Component Libraries](#component-libraries)\n- [Examples \u0026 Demos](#examples--demos)\n- [Articles \u0026 Talks](#articles--talks)\n- [Related Concepts](#related-concepts)\n\n---\n\n## What is Generative UI\n\n**Generative UI** is the practice of using AI models (typically LLMs) to dynamically generate user interface components at runtime, rather than having developers hand-code every interface.\n\n### The Spectrum\n\n| Dimension             | Traditional UI                  | AI-Assisted UI                        | Generative UI                             |\n| --------------------- | ------------------------------- | ------------------------------------- | ----------------------------------------- |\n| Primary workflow      | Developers hand-code components | Copilot helps developers write code   | Model generates UI components from intent |\n| Developer involvement | High (author everything)        | High (developer in the loop)          | Lower per UI (developer sets constraints) |\n| Runtime behavior      | Static UI                       | Static UI (generation happens in dev) | Dynamic / runtime-generated UI            |\n| Typical output        | Source code                     | Source code                           | Components, schemas, or rendered UI       |\n| Best for              | Predictability, control         | Faster delivery with human oversight  | Personalization and rapid UI composition  |\n\n### Why It Matters\n\nGenerative UI enables **personalization at scale**, **rapid prototyping**, and **natural language interfaces** (e.g., \"Show me sales by region\" produces an actual chart), while reducing development time by shifting effort from implementation details to intent.\n\n### Challenges\n\nChallenges include **Consistency** (generated UIs may look different each time), **Performance** (generation takes time; streaming helps), **Safety** (arbitrary code generation has security implications), and **Hallucination** (AI might generate components that don't exist).\n\n---\n\n## Key Concepts\n\n### Component Registry\nA predefined set of UI components the AI can use. Instead of generating arbitrary HTML/CSS, the AI selects from known, tested components.\n\n```\nRegistry: [BarChart, LineChart, DataTable, KPICard]\nPrompt: \"Show monthly revenue\"\nOutput: { component: \"BarChart\", props: { data: [...] } }\n```\n\n### Query Registry\nA predefined set of data sources the AI can query. Prevents hallucinated API calls.\n\n```\nRegistry: [getUsers, getRevenue, getOrders]\nPrompt: \"Show top customers\"\nOutput: { query: \"getUsers\", params: { sort: \"revenue\", limit: 10 } }\n```\n\n### Streaming UI\nRendering UI components as they're generated, rather than waiting for complete output.\n\n### Server Components\nReact Server Components enable streaming UI from server to client, making generative UI more practical.\n\n### Structured Output\nUsing JSON schemas or function calling to ensure AI outputs valid, parseable UI descriptions.\n\n### Constrained vs Unconstrained Generation\n\nA fundamental design decision in generative UI systems:\n\n| Approach          | Description                           | Examples                                                              |\n| ----------------- | ------------------------------------- | --------------------------------------------------------------------- |\n| **Constrained**   | AI selects from registered components | Vercel AI SDK `streamUI()`, A2UI, Crayon, assistant-ui, Tambo, v0.dev |\n| **Unconstrained** | AI generates raw HTML/CSS/JS directly | Google GenUI research, Claude Artifacts, MCP Apps, OpenAI Canvas      |\n\n**Constrained systems** are safer and more consistent — the AI can only use pre-approved components with known behavior. Trade-off: less flexible, requires component development upfront.\n\n**Unconstrained systems** are more powerful — the AI can create anything. Trade-off: needs sandboxing (iframe, E2B), potential for inconsistency, and security considerations.\n\nGoogle's research (see below) suggests unconstrained generation may become the dominant approach as models improve. They found users preferred AI-generated HTML/CSS over markdown 83% of the time, calling it an \"emergent capability\" — models produce good UIs without UI-specific training.\n\nIn practice, the 2025–2026 wave of production systems has leaned the other way. AG-UI, Google's own A2UI spec, the MCP Apps extension, and SDKs like Crayon and assistant-ui all standardize on *constrained / declarative* output — agents emit allow-listed components rather than raw code — for safety and consistency. Both ends of the spectrum are advancing in parallel rather than one cleanly winning.\n\n---\n\n## Research\n\n### Papers\n\n- [Generative UI: LLMs are Effective UI Generators](https://generativeui.github.io/static/pdfs/paper.pdf) - Google (2025). Foundational paper making the case for unconstrained UI generation. Key findings: (1) Users preferred AI-generated HTML/CSS/JS over markdown 83% of the time, (2) UI generation is an \"emergent capability\" requiring no UI-specific training, (3) Vision of \"infinite ephemeral interfaces\" - every query gets a custom UI. ([Project page](https://generativeui.github.io/)).\n- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://arxiv.org/abs/2307.13854) - CMU (2023). Framework for evaluating web agents, relevant to UI generation and interaction.\n- [Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding](https://arxiv.org/abs/2210.03347) - Google (2023). Vision-language model pretrained on web screenshots. Foundation for screenshot-to-code approaches.\n\n*Note: Generative UI is an emerging field. Much knowledge lives in blog posts, SDKs, and industry practice rather than academic papers.*\n\n---\n\n## Benchmarks \u0026 Datasets\n\n*Benchmarks for evaluating UI generation and datasets that enable screenshot-to-code, layout understanding, and web-agent interaction.*\n\n- [Design2Code](https://arxiv.org/abs/2403.03163) - Benchmark for converting designs/screenshots into front-end code (Stanford/Google, 2024).\n- [WebArena](https://github.com/web-arena-x/webarena) - Realistic web environment and benchmark for agents interacting with live websites.\n- [VisualWebArena](https://github.com/web-arena-x/visualwebarena) - Vision-grounded WebArena variant for UI understanding and interaction.\n- [Mind2Web](https://github.com/OSU-NLP-Group/Mind2Web) - Dataset and benchmark for generalist web agents grounded in real webpages.\n- [MiniWoB++](https://github.com/Farama-Foundation/miniwob-plusplus) - Standard suite of web UI interaction tasks used for agent evaluation.\n- [RICO](https://www.interactionmining.org/archive/rico) - Large-scale dataset of mobile app UIs (screens + view hierarchies).\n- [pix2code](https://arxiv.org/abs/1705.07962) - Early reference paper on UI screenshot-to-code generation (Tony Beltramelli, 2017).\n\n---\n\n## Specifications \u0026 Protocols\n\n### Open Specifications\n\n- [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) - Anthropic. Protocol for connecting AI models to external data sources and tools.\n- [Model Context Protocol (MCP) Servers](https://github.com/modelcontextprotocol/servers) - Reference implementations of MCP servers and tooling.\n- [MCP Apps (UI Extension)](https://github.com/modelcontextprotocol/ext-apps) - The official MCP extension for interactive UI: tools return UI resources that render in an iframe inside the host client. Builds on mcp-ui and the OpenAI Apps SDK ([announcement](https://blog.modelcontextprotocol.io/posts/2026-01-26-mcp-apps/)).\n- [OpenAI Apps SDK](https://developers.openai.com/apps-sdk) - OpenAI. Build apps that render interactive React UI inline in ChatGPT; extends MCP with a UI layer via the MCP Apps bridge.\n- [mcp-ui](https://github.com/idosal/mcp-ui) - Ido Salomon and Liad Yosef. Community SDK (TypeScript, Ruby, Python) for serving interactive UI over MCP; pioneered the pattern that fed the official MCP Apps spec.\n\n### Agent-to-UI Protocols\n\n*Protocols that connect agent backends to front ends and describe the UI that agents emit.*\n\n- [AG-UI (Agent-User Interaction Protocol)](https://github.com/ag-ui-protocol/ag-ui) - CopilotKit. Open, event-based protocol that streams agent activity and shared state to front ends over SSE. Adopted by Google, LangChain, AWS, Microsoft, Mastra, and PydanticAI.\n- [A2UI](https://github.com/google/a2ui) - Google. Declarative, streaming (JSONL) generative-UI spec: agents request allow-listed components that the client renders natively (Flutter, Angular, Lit). A data format, not executable code ([site](https://a2ui.org/)).\n\n### How the Layers Fit Together\n\nThese specs are largely complementary rather than competing — they standardize different layers of the same stack:\n\n| Layer                       | What it standardizes                          | Examples                                                 |\n| --------------------------- | --------------------------------------------- | -------------------------------------------------------- |\n| Transport / events          | How agent activity and state stream to the UI | AG-UI                                                    |\n| UI description format       | How the UI itself is described in the payload | A2UI (declarative components), MCP Apps (HTML resources) |\n| In-client rendering surface | Where generated UI renders in a host chat app | MCP Apps, OpenAI Apps SDK, mcp-ui                        |\n\n---\n\n## Frameworks \u0026 SDKs\n\n### Generative UI Frameworks\n\nPurpose-built for streaming AI-generated interfaces:\n\n- [Vercel AI SDK](https://sdk.vercel.ai/) - Vercel. TypeScript toolkit with `streamUI()` for server-streamed components and multi-provider support; v5 adds React/Vue/Svelte parity and SSE-based streaming.\n- [Tambo](https://github.com/tambo-ai/tambo) - Generative UI SDK for React, purpose-built for streaming AI-generated components.\n- [Hashbrown](https://github.com/liveloveapp/hashbrown) - Framework for building generative user interfaces in Angular and React.\n- [Cuttlekit](https://cuttlekit.com) - Fully generative UI framework, framework agnostic, optimised for performance and real-time UI generation.\n- [mdocUI](https://github.com/mdocui/mdocui) - Streaming generative UI using Markdoc `{% %}` tag syntax. Framework-agnostic core with React renderer, 24 theme-neutral components, and Zod schema validation.\n- [CopilotKit](https://github.com/CopilotKit/CopilotKit) - Full-stack framework for in-app agents and generative UI across React, Angular, mobile, and Slack; makers of the AG-UI protocol.\n- [assistant-ui](https://github.com/assistant-ui/assistant-ui) - TypeScript/React primitives for AI chat with a first-class generative-UI primitive that renders agent-described components from a consumer-provided allowlist.\n- [Thesys C1](https://www.thesys.dev/) - Thesys. OpenAI-compatible API that returns rendered UI instead of text, paired with the MIT-licensed Crayon React toolkit (built on Radix and shadcn/ui patterns).\n- [OpenUI Lang (Thesys)](https://github.com/thesysdev/openui) - MIT-licensed, streaming-first DSL and React runtime for model-generated UI; a compact, token-efficient alternative to emitting component JSON.\n\n### Supporting Libraries\n\nBuilding blocks for reliable generation:\n\n- [Instructor](https://github.com/jxnl/instructor) - Structured output extraction; useful with UI schemas for reliable generation.\n- [Mitosis](https://github.com/BuilderIO/mitosis) - Builder.io. Write components once and compile to React/Vue/Svelte/etc.\n\n---\n\n## Tools \u0026 Platforms\n\n### Component Generators\n\n*\"Describe a component, get code.\"*\n\n- [v0.dev](https://v0.dev/) - Vercel (commercial). Generate React/Tailwind components from prompts; known for shadcn/ui integration.\n- [OpenUI (Weights \u0026 Biases)](https://github.com/wandb/openui) - Describe UI in natural language and see it rendered live.\n- [openv0](https://github.com/raidendotai/openv0) - Open source v0 clone that generates React/Tailwind components from prompts.\n\n### App Builders\n\n*\"Describe an app, get a deployable project.\"*\n\n- [Bolt.new](https://bolt.new/) - StackBlitz (commercial). Generate full-stack applications in-browser with live preview.\n- [Lovable](https://lovable.dev/) - Generate and deploy web apps from descriptions (commercial).\n- [Cofounder](https://github.com/raidendotai/cofounder) - Open source full-stack app generator with generative UI, databases, and APIs.\n\n### Design-to-Code\n\n*\"Convert visuals to working code.\"*\n\n- [Screenshot-to-Code](https://github.com/abi/screenshot-to-code) - Convert screenshots or designs to HTML/React/Vue code.\n- [tldraw make-real](https://makereal.tldraw.com/) - Turn a wireframe into a working React component.\n- [Galileo AI](https://www.usegalileo.ai/) - Generate editable UI designs from text descriptions (commercial, Figma-compatible).\n- [Google Stitch](https://stitch.withgoogle.com/) - Google Labs. Generates web and mobile UI designs plus front-end code from prompts, screenshots, or sketches.\n- [Figma Make](https://www.figma.com/make/) - Figma. Prompt-to-UI inside Figma that builds working interfaces using your existing components and design system.\n- [Magic Patterns](https://www.magicpatterns.com/) - Generate and iterate on React UI from prompts; can learn and apply an existing design system.\n- [Subframe](https://www.subframe.com/) - AI design tool built for code: compose production-ready React/Tailwind components on a canvas and export clean code.\n\n### MCP Tools for UI\n\n*\"Model Context Protocol servers that improve AI UI generation.\"*\n\n- [Figma-Context-MCP](https://github.com/GLips/Figma-Context-MCP) - GLips. Provides Figma layout information to AI coding agents.\n- [21st.dev Magic MCP](https://github.com/21st-dev/magic-mcp) - 21st.dev. Generate UI components from prompts with design-system awareness.\n- [shadcn-ui-mcp-server](https://github.com/Jpisnice/shadcn-ui-mcp-server) - Jpisnice. Helps LLMs understand shadcn/ui component structure.\n- [MUI MCP](https://mui.com/x/introduction/mcp/) - MUI. MCP server for MUI docs and code examples, published as `@mui/mcp` ([npm](https://www.npmjs.com/package/@mui/mcp)).\n- [Storybook MCP](https://github.com/storybookjs/mcp) - Storybook. MCP server and addon that exposes component information and workflows from your local Storybook ([npm](https://www.npmjs.com/package/@storybook/mcp)).\n- [Cursor Talk to Figma MCP](https://github.com/grab/cursor-talk-to-figma-mcp) - Grab. MCP server + Figma plugin for reading and modifying Figma designs.\n- [Playwright MCP](https://github.com/microsoft/playwright-mcp) - Microsoft. MCP server for browser automation and UI regression testing workflows.\n- [shadcn-vue-mcp](https://github.com/HelloGGX/shadcn-vue-mcp) - HelloGGX. MCP server for shadcn-vue component knowledge.\n\n### AI Products with UI Generation\n\n*\"Chat interfaces that can generate UIs.\"*\n\n- [Claude](https://claude.ai/) - Anthropic (commercial). Artifacts generates interactive React/HTML UIs in chat.\n- [ChatGPT](https://chatgpt.com/) - OpenAI (commercial). Canvas supports UI generation and editing.\n\n---\n\n## Security \u0026 Sandboxing\n\n*When UIs (or UI code) are model-generated, treat outputs as untrusted. These resources cover browser sandboxing, sanitization, and LLM-app security pitfalls.*\n\n- [OWASP Top 10 for LLM Applications](https://owasp.org/www-project-top-10-for-large-language-model-applications/) - OWASP. Threat model checklist for LLM apps (prompt injection, insecure tool use, data leakage).\n- [OWASP Prompt Injection Prevention Cheat Sheet](https://cheatsheetseries.owasp.org/cheatsheets/LLM_Prompt_Injection_Prevention_Cheat_Sheet.html) - OWASP. Practical mitigations for prompt/tool injection in production systems.\n- [Content Security Policy (CSP)](https://developer.mozilla.org/en-US/docs/Web/HTTP/Guides/CSP) - MDN. Core browser control for limiting script execution and exfiltration.\n- [iframe sandbox](https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/Elements/iframe) - MDN. Reference for sandboxed iframes to contain untrusted HTML/JS UI previews.\n- [Trusted Types](https://web.dev/articles/trusted-types) - Mitigation against DOM XSS when inserting generated HTML into the DOM.\n- [DOMPurify](https://github.com/cure53/DOMPurify) - Widely used HTML sanitizer for untrusted/generated markup.\n- [Sandpack](https://github.com/codesandbox/sandpack) - CodeSandbox. In-browser code sandboxing patterns for safe previews.\n- [E2B](https://github.com/e2b-dev/e2b) - Sandboxed code execution environments for running untrusted generated code.\n\n---\n\n## Evaluation \u0026 Testing\n\n*Tools and practices for regression testing generated UIs, validating structured outputs, and red-teaming LLM apps.*\n\n- [Playwright](https://playwright.dev/) - Microsoft. E2E testing and screenshot diffs for UI regression testing.\n- [Storybook Test Runner](https://storybook.js.org/docs/writing-tests/integrations/test-runner) - Storybook. Automates component-level interaction testing.\n- [promptfoo](https://github.com/promptfoo/promptfoo) - Prompt and tool-call regression testing across models.\n- [PyRIT](https://github.com/Azure/PyRIT) - Microsoft. Red teaming toolkit for LLM apps (jailbreaks, prompt injection).\n- [garak](https://github.com/NVIDIA/garak) - NVIDIA. LLM vulnerability scanner for automated probing.\n\n---\n\n## Schemas \u0026 Patterns\n\n*Make model outputs reliable: constrain generation, validate payloads, and standardize tool/component contracts.*\n\n- [JSON Schema](https://json-schema.org/) - Standard for validating structured model outputs.\n- [OpenAPI Specification](https://spec.openapis.org/oas/latest.html) - OpenAPI Initiative. Standard contract format for APIs and tool catalogs.\n- [Zod](https://github.com/colinhacks/zod) - Colin Hacks. TypeScript schema validation for enforcing output shapes at runtime.\n- [Outlines](https://github.com/dottxt-ai/outlines) - Structured generation with JSON/grammar constraints.\n- [Guidance](https://github.com/guidance-ai/guidance) - Grammar- and constraint-oriented prompting for structured outputs.\n- [Vercel AI SDK](https://sdk.vercel.ai/docs) - Vercel. Streaming UI and tool-calling patterns for generative UI applications.\n- [shadcn/ui Registry](https://ui.shadcn.com/docs/registry) - Component registry pattern commonly targeted by UI generators.\n\n---\n\n## Component Libraries\n\n### Generation Targets\n\n*Libraries that AI generators commonly output to:*\n\n- [shadcn/ui](https://ui.shadcn.com/) - Copy-paste components that many generators target.\n- [Radix UI](https://www.radix-ui.com/) - Radix. Unstyled primitives that shadcn/ui builds on.\n- [Tremor](https://tremor.so/) - Dashboard components often used for generated data UIs.\n- [React Native Paper](https://reactnativepaper.com/) - Material Design components for React Native.\n\n### For Building AI/Chat Interfaces\n\n*Components designed for LLM-powered apps:*\n\n- [AI Elements](https://vercel.com/changelog/introducing-ai-elements) - Vercel. 20+ shadcn/ui-based React components for AI interfaces (message threads, reasoning panels, tool output), integrated with the AI SDK.\n- [LangUI](https://github.com/LangbaseInc/langui) - LangbaseInc. Tailwind components for chat, AI assistants, and LLM projects.\n- [GPT-Vis](https://github.com/antvis/GPT-Vis) - AntV. Visualization components designed for LLM-generated outputs.\n\n### Visualization\n\n*For AI-generated charts and data displays:*\n\n- [Recharts](https://recharts.org/) - Composable React charts commonly used in dashboards.\n- [Vega-Lite](https://vega.github.io/vega-lite/) - Vega. Declarative JSON grammar for charts and visualizations.\n- [Observable Plot](https://observablehq.com/plot/) - Observable. Grammar of graphics for JS with concise specs.\n- [D3.js](https://d3js.org/) - D3. Low-level but powerful library for custom visualizations.\n\n---\n\n## Examples \u0026 Demos\n\n### Open Source\n\n- [ai-chatbot](https://github.com/vercel/ai-chatbot) - Vercel. Full-featured chatbot with generative UI using the AI SDK.\n- [morphic](https://github.com/miurla/morphic) - AI-powered search engine with generative answer UI.\n- [langgraphjs-gen-ui-examples](https://github.com/langchain-ai/langgraphjs-gen-ui-examples) - LangChain. Collection of generative UI agents built with LangGraph.js.\n- [gen-ui](https://github.com/bracesproul/gen-ui) - Reference implementation for generative UI with LangChain.js, AI SDK, and Next.js.\n- [termite](https://github.com/shobrook/termite) - Terminal-based generative UI.\n\n---\n\n## Articles \u0026 Talks\n\n- [Vercel: Introducing AI SDK 3.0 with Generative UI](https://vercel.com/blog/ai-sdk-3-generative-ui) - Vercel blog post introducing Generative UI features in the AI SDK.\n- [Building effective agents](https://www.anthropic.com/engineering/building-effective-agents) - Anthropic engineering post on agent design patterns relevant to tool-using UI systems.\n- [Introducing A2UI: An open project for agent-driven interfaces](https://developers.googleblog.com/introducing-a2ui-an-open-project-for-agent-driven-interfaces/) - Google Developers blog introducing the A2UI declarative generative-UI spec.\n- [AG-UI Protocol: Bridging Agents to Any Front End](https://www.copilotkit.ai/blog/ag-ui-protocol-bridging-agents-to-any-front-end) - CopilotKit post explaining the agent-to-frontend interaction protocol.\n\n---\n\n## Related Concepts\n\n### Adjacent Fields\n\nDesign-to-code converts designs (e.g., Figma) to code. Code generation is the broader field of AI-generated code. Conversational UI covers chat interfaces (related but distinct). Low-code/no-code includes visual builders that generative UI can automate.\n\n- [LangChain.js](https://js.langchain.com/) - LangChain. LLM framework often used to build tool-using agents that can drive generative UI.\n- [Mastra](https://mastra.ai/) - TypeScript framework for AI applications with patterns adaptable to generative UI.\n\n### Evaluation tooling (adjacent)\n\n- [OpenAI Evals](https://github.com/openai/evals) - OpenAI. Framework for evaluating model outputs with custom tasks and graders.\n- [TruLens](https://github.com/truera/trulens) - TruEra. Evaluation and feedback tooling for LLM applications.\n- [LangSmith Evaluation](https://docs.smith.langchain.com/evaluation) - LangChain. Evaluation patterns and workflows for LLM apps.\n\n### Standards \u0026 Formats\n\nJSON Schema defines valid UI structures. OpenAPI describes APIs for query registries. Web Components are a framework-agnostic component standard.\n\n---\n\n## Contributing\n\nContributions welcome! Please read the [contribution guidelines](CONTRIBUTING.md) first.\n\n### Criteria for Inclusion\n\nSubmissions should be relevant to generative UI (not general AI/LLM), high quality (well-documented and maintained or historically significant), and accessible (open source, free tier, or detailed documentation).\n\n### What We're Looking For\n\nNew frameworks and tools, research papers and articles, interesting examples and demos, and corrections/updates.\n\n---\n\nLicense: CC BY 4.0 (see [LICENSE](LICENSE)).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnarrowin%2Fawesome-generative-ui","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnarrowin%2Fawesome-generative-ui","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnarrowin%2Fawesome-generative-ui/lists"}