awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
CopilotKit avatar

CopilotKit/OpenIntelligentUI

0
View on GitHub↗
opengenerativeui.copilotkit.ai↗

OpenIntelligentUI

OpenIntelligentUI is a generative UI chat application and forkable starter in which an AI agent answers questions with live, interactive interface elements instead of plain text. Each chat message starts an agent run that chooses between a plain text reply, a prebuilt native component, or a custom generated interface, and the result streams into the conversation as it is produced.

What sets it apart is how it handles generated interfaces. A routing step reads each turn and picks a renderer, using tables for simple data and charts, diagrams, calculators, or maps for richer requests. Agent-written HTML, CSS, and JavaScript stream into an isolated iframe, so live interfaces run without touching the host page. Values chosen inside the sandbox return to the host through a validated message bridge, which can start a follow-up agent turn. The project also includes a skill server that exposes skills, prompt templates, and an HTML document assembler as tools over a standard protocol, reachable via HTTP, stdio, or Docker.

Users bring their own model and visualization-provider keys. The app validates them with a test request before saving, holds them only in browser memory until refresh, keeps them out of chat state and checkpoints, and bills requests to the user's own account. Provider errors are reported to the user rather than silently switching providers.

The generated output covers working calculators such as bill splitters, side-by-side comparisons that update with preferences, interactive trip maps with numbered pins, rotatable 3D models, and tables convertible into charts. The chat view preserves scroll position while streaming and offers a copy action. The model is configured through environment settings, a health endpoint confirms the agent service is running, and the interface and agent can be modified after cloning.

Features

  • Agentic UI Frameworks - Lets an AI agent answer with generated live interface components, choosing between text, native components, or custom interfaces for each request.
  • Generative User Interfaces - Turn user questions into live interface elements such as explanations, comparisons, and working tools, instead of returning plain text responses.
  • Interactive Agent Chat Interfaces - Opens directly to a chat view where typed requests or starter suggestions start an agent run and render text and interactive results inline.
  • Request-to-Output Format Routing - Selects the renderer and visualization type per user turn, using tables for simple data and charts, diagrams, calculators, or maps for richer requests.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
4,185 stars·437 forks·TypeScript·MIT·1 view
  • Rich Answer Chat Interfaces - Delivers a chat interface where agent answers appear as live interactive components and text, rendered progressively as the response streams in.
  • Generated Task Tools - Creates working calculators, planners, and interactive models with functioning controls on demand for the user's immediate task.
  • Double-Iframe Renderers - Streams agent-written HTML, CSS, and JavaScript into an isolated iframe so live interfaces run without touching the host page's code or data.
  • Generative UI Components - Stream HTML, CSS, and JavaScript from an agent into an isolated iframe, so generated interfaces display live without touching the host application.
  • Browser Agent Chat Clients - Provides a browser chat client that streams agent output and renders text and interactive results inline as the agent runs.
  • Sandboxed Generated Page Rendering - Runs agent-written HTML, CSS, and JavaScript inside an isolated iframe so generated interfaces work live without touching the host page.
  • Agent Response Streamers - Streams agent output to the browser incrementally so text and interface elements appear progressively while the answer is still being written.
  • Bring-Your-Own-Key Model Clients - Lets users supply their own model and provider keys, validated before saving and held only in browser memory, so costs go to their accounts.
  • Bring-Your-Own-Key Model Access - Collects users' own model and visualization-provider keys in the chat interface so requests bill to their accounts instead of shared credentials.
  • Chat-Triggered Agent Runs - Starts an agent run for every chat message that picks between plain text, a native component, or a custom interface and streams it inline.
  • Visualization Routers - Routes each user turn to a table, chart, diagram, calculator, or map renderer depending on how rich the request is.
  • Model Context Protocol Servers - Exposes skills, prompt templates, and an HTML document assembler as tools over the Model Context Protocol via HTTP, stdio, or Docker.
  • Agent Response Streams - Streams the agent's response to the web front end as it is produced so generated interface elements appear progressively.
  • Interactive Calculators and Planners - Generates working tools such as bill splitters and comparison views on demand, with results updating instantly as inputs change.
  • Interactive Data and Concept Visualizations - Turning questions into charts, tables, diagrams, maps, or rotatable 3D models so users can explore data and understand mechanisms by adjusting variables.
  • User-Provided Keys - Keeps user-supplied provider keys in browser memory only, out of chat state and checkpoints, cleared on refresh, billing the user's own account.
  • Sandbox Message Bridges - Passes values chosen inside the sandbox to the host through a validated message bridge that can start follow-up agent turns safely.
  • Sandbox-to-Host Message Bridges - Pass values selected inside the sandbox to the host through a validated bridge, starting another agent turn without exposing the host to untrusted code.
  • Session-Only Credential Handling - Hold user-entered keys in browser memory until refresh or clear, exclude them from chat state and checkpoints, and disable hosted tracing for those requests.
  • Iframe-Based Rendering - Streams agent-written HTML, CSS, and JavaScript into an isolated iframe, keeping generated interfaces separate from the host application.
  • Interactive Calculators - Generates working calculators such as bill splitters with total, tip, and group-size inputs that recalculate instantly as inputs change.
  • Prebuilt UI Components - Map supported structured tasks to prebuilt native components, so common requests display consistent interface elements instead of freshly generated markup.
  • Answer Format Selection - Choose between a plain text reply, a native component, or a custom interactive interface based on what best helps the user finish their request.

    • Star history

    Star history chart for copilotkit/openintelligentuiStar history chart for copilotkit/openintelligentui

    How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

    Projects sharing features with OpenIntelligentUI

    These projects share indexed features with OpenIntelligentUI. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
    • thunderbird/thunderboltthunderbird avatar

      thunderbird/thunderbolt

      4,819View on GitHub↗

      Thunderbolt is an open-source, self-hostable AI chat client that runs on web, desktop (macOS, Linux, Windows) and mobile (iOS, Android) from one shared web codebase inside lightweight native shells. It is local-first: each device keeps a local database as the source of truth, and data syncs to a central backend with optional end-to-end encryption, so the server stores only ciphertext for protected content. New devices are approved from trusted ones by fingerprint, recovery phrase or passkey, and revoked devices lose access. What sets it apart is its approach to model and agent choice and to p

      TypeScriptaiai-agentsllms
      View on GitHub↗4,819
    • mervinpraison/praisonaiMervinPraison avatar

      MervinPraison/PraisonAI

      9,325View on GitHub↗

      PraisonAI is an autonomous AI agent platform that coordinates multiple LLM-powered agents for research, planning, and execution of complex workflows. It functions as a multi-agent orchestration framework, a workflow builder, and a Model Context Protocol server, while also providing retrieval-augmented generation through vector knowledge bases. Agents can interact via CLI, web, or standardized protocols with sandboxed code execution. The platform distinguishes itself with a rich set of agent communication protocols, including A2A, REST, WebSocket, voice and telephony integration, and MCP, allo

      Pythonagentsaiai-agent-framework
      View on GitHub↗9,325
    • prefecthq/fastmcpPrefectHQ avatar

      PrefectHQ/fastmcp

      22,994View on GitHub↗

      FastMCP is a Python framework designed for building servers that expose functions, resources, and prompts to AI models using the Model Context Protocol. It simplifies the development process by automatically deriving tool metadata, input schemas, and documentation directly from Python function signatures and type hints. The framework provides a unified container for managing these components, allowing developers to build modular applications that integrate seamlessly with AI assistants. The project distinguishes itself through its support for interactive, server-defined user interface compone

      Pythonagentsfastmcpllms
      View on GitHub↗22,994
    • vercel/aivercel avatar

      vercel/ai

      21,885View on GitHub↗

      This project is a comprehensive framework for building AI-powered applications, providing a unified toolkit for orchestrating language models, autonomous agents, and interactive user interfaces. It serves as a central library for managing the entire lifecycle of AI interactions, from initial prompt generation and model provider abstraction to complex, multi-step reasoning and tool execution. The framework distinguishes itself through its deep integration with frontend development, specifically by enabling generative user interfaces that render dynamic components directly from model outputs. I

      TypeScriptanthropicartificial-intelligencegemini
      View on GitHub↗21,885
    Compare all 30 related projects→

    Frequently asked questions

    What does copilotkit/openintelligentui do?

    OpenIntelligentUI is a generative UI chat application and forkable starter in which an AI agent answers questions with live, interactive interface elements instead of plain text. Each chat message starts an agent run that chooses between a plain text reply, a prebuilt native component, or a custom generated interface, and the result streams into the conversation as it is produced.

    What are the main features of copilotkit/openintelligentui?

    The main features of copilotkit/openintelligentui are: Agentic UI Frameworks, Generative User Interfaces, Interactive Agent Chat Interfaces, Request-to-Output Format Routing, Rich Answer Chat Interfaces, Generated Task Tools, Double-Iframe Renderers, Generative UI Components.

    Which projects share features with copilotkit/openintelligentui?

    Projects with overlapping indexed features include: thunderbird/thunderbolt — Thunderbolt is an open-source, self-hostable AI chat client that runs on web, desktop (macOS, Linux, Windows) and… mervinpraison/praisonai — PraisonAI is an autonomous AI agent platform that coordinates multiple LLM-powered agents for research, planning, and… prefecthq/fastmcp — FastMCP is a Python framework designed for building servers that expose functions, resources, and prompts to AI models… vercel/ai — This project is a comprehensive framework for building AI-powered applications, providing a unified toolkit for… mcp-ui-org/mcp-ui — mcp-ui is a toolkit and framework for rendering interactive web components and managing communication between Model… anthropics/claude-agent-sdk-python — Claude Agent SDK is a Python library from Anthropic for building AI agents that use Claude’s tool‑calling, streaming,…