You can let the copilot interact with your app via plain typescript closures, and give it (explicit) read access to app data.
An example user interaction could look like:\n- "Which of these travel destinations has a rich architecture history?"\n- [Copilot answers]\n- "Great. Add these to my august trip folder, except the ones where it's typically rainy then"\n- [Added]
Recursive agent integration (via Langchain) is in the works - if you have ideas I'd love to hear them here or on the discord."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Show HN: CopilotKit \u2013 a hackable OSS copilot for any react app"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/RecursivelyAI/CopilotKit"}},"_tags":["story","author_swiftlyTyped","story_36817991","show_hn"],"author":"swiftlyTyped","children":[36818044,36821433],"created_at":"2023-07-21T18:59:18Z","created_at_i":1689965958,"num_comments":2,"objectID":"36817991","points":10,"story_id":36817991,"story_text":"CopilotKit is a typescript library for adding a hackable copilot to any react app.
You can let the copilot interact with your app via plain typescript closures, and give it (explicit) read access to app data.
An example user interaction could look like:\n- "Which of these travel destinations has a rich architecture history?"\n- [Copilot answers]\n- "Great. Add these to my august trip folder, except the ones where it's typically rainy then"\n- [Added]
Recursive agent integration (via Langchain) is in the works - if you have ideas I'd love to hear them here or on the discord.","title":"Show HN: CopilotKit \u2013 a hackable OSS copilot for any react app","updated_at":"2024-09-20T14:34:50Z","url":"https://github.com/RecursivelyAI/CopilotKit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"swiftlyTyped"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit raises $27M to build the Agentic FrontEnd Stack"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://techcrunch.com/2026/05/05/copilotkit-raises-27m-to-help-devs-deploy-app-native-ai-agents/"}},"_tags":["story","author_swiftlyTyped","story_48037265"],"author":"swiftlyTyped","created_at":"2026-05-06T15:19:53Z","created_at_i":1778080793,"num_comments":0,"objectID":"48037265","points":9,"story_id":48037265,"title":"CopilotKit raises $27M to build the Agentic FrontEnd Stack","updated_at":"2026-05-06T15:35:41Z","url":"https://techcrunch.com/2026/05/05/copilotkit-raises-27m-to-help-devs-deploy-app-native-ai-agents/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"nathan_tarbert"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Special feature: building agentic apps with CopilotKit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://www.buildingaiagents.ai/p/special-feature-building-agentic-apps-with-copilotkit"}},"_tags":["story","author_nathan_tarbert","story_42807251"],"author":"nathan_tarbert","children":[42807596],"created_at":"2025-01-23T19:36:15Z","created_at_i":1737660975,"num_comments":0,"objectID":"42807251","points":7,"story_id":42807251,"title":"Special feature: building agentic apps with CopilotKit","updated_at":"2025-01-28T15:14:30Z","url":"https://www.buildingaiagents.ai/p/special-feature-building-agentic-apps-with-copilotkit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ataitawkit"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit is comprehensive copilot infrastructure for react apps. Fully open-source.
Just added CopilotTextarea - a drop-in replacement for <textarea/> that supports context-aware copilot-style autocompletions with any LLM/backend.
Also coming up for CopilotTextarea (VERY soon): rich text support, and cursor-style insertions/edits."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Show HN: CopilotKit adds CopilotTextarea (built with slateJS)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/RecursivelyAI/CopilotKit"}},"_tags":["story","author_ataitawkit","story_37346134","show_hn"],"author":"ataitawkit","children":[37346396,37346536],"created_at":"2023-09-01T02:40:03Z","created_at_i":1693536003,"num_comments":3,"objectID":"37346134","points":5,"story_id":37346134,"story_text":"CopilotKit is comprehensive copilot infrastructure for react apps. Fully open-source.
Just added CopilotTextarea - a drop-in replacement for <textarea/> that supports context-aware copilot-style autocompletions with any LLM/backend.
Also coming up for CopilotTextarea (VERY soon): rich text support, and cursor-style insertions/edits.","title":"Show HN: CopilotKit adds CopilotTextarea (built with slateJS)","updated_at":"2024-09-20T15:01:25Z","url":"https://github.com/RecursivelyAI/CopilotKit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"psyphy"},"title":{"fullyHighlighted":true,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/RecursivelyAI/CopilotKit"}},"_tags":["story","author_psyphy","story_36808520"],"author":"psyphy","created_at":"2023-07-21T00:13:35Z","created_at_i":1689898415,"num_comments":0,"objectID":"36808520","points":5,"story_id":36808520,"title":"CopilotKit","updated_at":"2024-09-20T14:33:45Z","url":"https://github.com/RecursivelyAI/CopilotKit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"yarapavan"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit: The Open-Source Copilot Framework"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/CopilotKit/CopilotKi"}},"_tags":["story","author_yarapavan","story_40905893"],"author":"yarapavan","children":[40906103],"created_at":"2024-07-08T14:44:45Z","created_at_i":1720449885,"num_comments":1,"objectID":"40905893","points":4,"story_id":40905893,"title":"CopilotKit: The Open-Source Copilot Framework","updated_at":"2024-09-20T17:21:49Z","url":"https://github.com/CopilotKit/CopilotKi"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"handfuloflight"},"title":{"fullyHighlighted":true,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://www.copilotkit.ai/"}},"_tags":["story","author_handfuloflight","story_40531871"],"author":"handfuloflight","created_at":"2024-05-31T05:44:49Z","created_at_i":1717134289,"num_comments":0,"objectID":"40531871","points":3,"story_id":40531871,"title":"CopilotKit","updated_at":"2024-09-20T17:10:21Z","url":"https://www.copilotkit.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ulidabess"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Add Plug and Play Copilot Features to Any React App \u2013 CopilotKit (Open Source)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/RecursivelyAI/CopilotKit"}},"_tags":["story","author_ulidabess","story_37826776"],"author":"ulidabess","children":[37826777],"created_at":"2023-10-09T23:31:33Z","created_at_i":1696894293,"num_comments":1,"objectID":"37826776","points":2,"story_id":37826776,"title":"Add Plug and Play Copilot Features to Any React App \u2013 CopilotKit (Open Source)","updated_at":"2024-09-20T15:21:36Z","url":"https://github.com/RecursivelyAI/CopilotKit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"lio-p"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Building an agentic app with ClickHouse MCP and CopilotKit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://clickhouse.com/blog/building-an-agentic-application-with-clickhouse-mcp-server-and-copilotkit"}},"_tags":["story","author_lio-p","story_44289381"],"author":"lio-p","children":[44289388],"created_at":"2025-06-16T13:36:08Z","created_at_i":1750080968,"num_comments":0,"objectID":"44289381","points":1,"story_id":44289381,"title":"Building an agentic app with ClickHouse MCP and CopilotKit","updated_at":"2025-06-16T13:40:34Z","url":"https://clickhouse.com/blog/building-an-agentic-application-with-clickhouse-mcp-server-and-copilotkit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"chkelly9"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Building an agentic app with ClickHouse MCP and CopilotKit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://clickhouse.com/blog/building-an-agentic-application-with-clickhouse-mcp-server-and-copilotkit"}},"_tags":["story","author_chkelly9","story_44261476"],"author":"chkelly9","created_at":"2025-06-12T18:45:54Z","created_at_i":1749753954,"num_comments":0,"objectID":"44261476","points":1,"story_id":44261476,"title":"Building an agentic app with ClickHouse MCP and CopilotKit","updated_at":"2025-06-12T18:49:29Z","url":"https://clickhouse.com/blog/building-an-agentic-application-with-clickhouse-mcp-server-and-copilotkit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"yousefh409"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Hi HN, we\u2019re Yousef & Nour, founders of Vendo (https://vendo.run). Vendo lets users create new features inside the software they already use. A user describes the dashboard, workflow, or small app they need, and Vendo builds it on top of the product\u2019s existing data, API, and interface.
Demo: https://www.youtube.com/watch?v=VdpHehY64ls
We built Vendo because every SaaS eventually faces the same problem: every customer needs something slightly different. One wants a new report and another needs a workflow that only makes sense for their team. These requests either sit on the roadmap, become one-off engineering work, or force the customer into spreadsheets and external tools. We wanted the user to be able to create the missing feature themselves, without leaving the product.
Here is how it works:
- npx vendo init reads the product's API surface, theme, routes, and more. These are used so that the apps Vendo creates (1) look on-brand and native and (2) have the ability to read data and perform actions directly through the company's API
- When a user asks for a feature, we have a custom Vendo harness that writes a React component with a bunch of Vendo add-ons and guardrails (ex. ability to make calls to the host API + our component library). Every save is compiled, type-checked, run against real API responses, and rendered before the user sees it. We just released a benchmark and write-up here with more info for anyone interested: https://vendo.run/blog/generating-product-ui-measured
- We use QuickJS to make sure that anything the agent creates is sandboxed and can't mess with the company's site. Vendo compiles the component and runs it with Preact inside a QuickJS VM with no access to the DOM, network, or clock. The VM returns a UI tree, which the host renders using the product\u2019s registered components. When the user clicks something, QuickJS emits a tool call; the host executes it through Vendo\u2019s guard and passes the result back into the same VM, preserving the screen\u2019s local state.
There's a lot of generative UI right now: streaming developer-written components into a chat (Vercel AI SDK, CopilotKit, Thesys), or rendering your app inside someone else's assistant (OpenAI Apps SDK, MCP Apps). We differ on two things. Vendo lives in your product and acts through your API as the signed-in user, so what it makes is durable: real apps users keep, pin, and run on triggers while they're away, and not components that are merely confined to a chat. Plus, it's not capped at putting together a bunch of prebuilt components: the agent can build arbitrary apps, from a quick dashboard out of your own components to real custom code running in a sandbox, and either way data only ever comes from tool calls to your API.
Here are some things customers are using Vendo for today:
- Letting their users create custom dashboards and reports. These are mainly UI-based and focused on letting the user see the exact graphs and metrics they care about
- Letting their customers create recurring automations. A big thing as well that has been used for these automations is the fact that we connect to external connections, so users have been automating many of their inter-tool workflows (ex. an automation that sends a slack alert based off of something in the product)
- B2B customers letting their customers customize the product with specific business logic. Often this is simple things like an extra field on a form, or an extra permission, but it is hard for a business to keep up with them otherwise.
- Creating and sharing custom dashboards/apps across an organization. Since the apps Vendo creates are durable, they can be shared, reused, and forked (which can\u2019t be done with many of the other in-chat generative UI solutions)
We've spent a lot of time thinking about how AI and agents will change the way people consume software. We think the answer is personal(ized) software: you see the UI you need to see, you tell an agent exactly what you need, and the product molds to how you work.
The key insights that have enabled the product to work are:
- A rule in code always beats a rule in a prompt.
- Invent as little syntax as possible. Generation got faster and more reliable when the output looked like what models already know (JSX-shaped markup) instead of a clever custom format.
- Deterministic beats model wherever you can get away with it. Theme extraction is pure static analysis, and a remix starts as a copy of your component, no model call.
Vendo is completely open-source (Apache-2.0) and can be self-hosted, so feel free to check out all the source code here: https://github.com/runvendo/vendo.
Would love you to try it out and give us your feedback: https://docs.vendo.run/. Or if you\u2019re a company looking to embed Vendo in your product feel free to book a call here: https://cal.com/team/vendo/intro-call"},"title":{"matchLevel":"none","matchedWords":[],"value":"Launch HN: Vendo (YC S26) \u2013 Let users build features on top of your product"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/runvendo/vendo"}},"_tags":["story","author_yousefh409","story_49376038","launch_hn"],"author":"yousefh409","children":[49376394,49376770,49377408,49378319,49378626,49378955,49378974,49384204,49385351,49388187,49388425],"created_at":"2026-08-20T15:29:52Z","created_at_i":1787239792,"num_comments":22,"objectID":"49376038","points":61,"story_id":49376038,"story_text":"Hi HN, we\u2019re Yousef & Nour, founders of Vendo (https://vendo.run). Vendo lets users create new features inside the software they already use. A user describes the dashboard, workflow, or small app they need, and Vendo builds it on top of the product\u2019s existing data, API, and interface.
Demo: https://www.youtube.com/watch?v=VdpHehY64ls
We built Vendo because every SaaS eventually faces the same problem: every customer needs something slightly different. One wants a new report and another needs a workflow that only makes sense for their team. These requests either sit on the roadmap, become one-off engineering work, or force the customer into spreadsheets and external tools. We wanted the user to be able to create the missing feature themselves, without leaving the product.
Here is how it works:
- npx vendo init reads the product's API surface, theme, routes, and more. These are used so that the apps Vendo creates (1) look on-brand and native and (2) have the ability to read data and perform actions directly through the company's API
- When a user asks for a feature, we have a custom Vendo harness that writes a React component with a bunch of Vendo add-ons and guardrails (ex. ability to make calls to the host API + our component library). Every save is compiled, type-checked, run against real API responses, and rendered before the user sees it. We just released a benchmark and write-up here with more info for anyone interested: https://vendo.run/blog/generating-product-ui-measured
- We use QuickJS to make sure that anything the agent creates is sandboxed and can't mess with the company's site. Vendo compiles the component and runs it with Preact inside a QuickJS VM with no access to the DOM, network, or clock. The VM returns a UI tree, which the host renders using the product\u2019s registered components. When the user clicks something, QuickJS emits a tool call; the host executes it through Vendo\u2019s guard and passes the result back into the same VM, preserving the screen\u2019s local state.
There's a lot of generative UI right now: streaming developer-written components into a chat (Vercel AI SDK, CopilotKit, Thesys), or rendering your app inside someone else's assistant (OpenAI Apps SDK, MCP Apps). We differ on two things. Vendo lives in your product and acts through your API as the signed-in user, so what it makes is durable: real apps users keep, pin, and run on triggers while they're away, and not components that are merely confined to a chat. Plus, it's not capped at putting together a bunch of prebuilt components: the agent can build arbitrary apps, from a quick dashboard out of your own components to real custom code running in a sandbox, and either way data only ever comes from tool calls to your API.
Here are some things customers are using Vendo for today:
- Letting their users create custom dashboards and reports. These are mainly UI-based and focused on letting the user see the exact graphs and metrics they care about
- Letting their customers create recurring automations. A big thing as well that has been used for these automations is the fact that we connect to external connections, so users have been automating many of their inter-tool workflows (ex. an automation that sends a slack alert based off of something in the product)
- B2B customers letting their customers customize the product with specific business logic. Often this is simple things like an extra field on a form, or an extra permission, but it is hard for a business to keep up with them otherwise.
- Creating and sharing custom dashboards/apps across an organization. Since the apps Vendo creates are durable, they can be shared, reused, and forked (which can\u2019t be done with many of the other in-chat generative UI solutions)
We've spent a lot of time thinking about how AI and agents will change the way people consume software. We think the answer is personal(ized) software: you see the UI you need to see, you tell an agent exactly what you need, and the product molds to how you work.
The key insights that have enabled the product to work are:
- A rule in code always beats a rule in a prompt.
- Invent as little syntax as possible. Generation got faster and more reliable when the output looked like what models already know (JSX-shaped markup) instead of a clever custom format.
- Deterministic beats model wherever you can get away with it. Theme extraction is pure static analysis, and a remix starts as a copy of your component, no model call.
Vendo is completely open-source (Apache-2.0) and can be self-hosted, so feel free to check out all the source code here: https://github.com/runvendo/vendo.
Would love you to try it out and give us your feedback: https://docs.vendo.run/. Or if you\u2019re a company looking to embed Vendo in your product feel free to book a call here: https://cal.com/team/vendo/intro-call","title":"Launch HN: Vendo (YC S26) \u2013 Let users build features on top of your product","updated_at":"2026-08-22T16:42:35Z","url":"https://github.com/runvendo/vendo"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"becomevocal"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Hey everyone. My cofounder and I are open sourcing Persona.js (https://www.persona-chat.dev/). It's a VanillaJS library that helps anyone build agentic experiences on the web, without a framework dependency, and full WebMCP support
So, why'd we do this?
1) We're super fans of the web and the browser can do a ton today
2) We've seen AI builds be way overly complex because the FE requires a large project within an existing app OR the site wasn't using a framework to begin with
If you've been a part of huge apps with multiple frameworks inside them, or work inside CMS / website builders / ecomm platforms... you know what we're talking about. A "simple" AI feature disrupts your life for months
If you have a singular React codebase and can't possibly imagine building an interface without JSX, nothing to see here! You already have a lot of great options that are really cool too. Check out Assistant UI / CopilotKit / AI Elements which are all MIT.
There's a demo video here: https://www.youtube.com/watch?v=68D80uNsfH0.
Some specifics:
Persona has a ~15 kB brotli to first paint (the full widget lazy-loads on first click) while being able to render most of the primary agent experiences you see on the web: from 'Fin' (pill launcher) to 'Claude' (fullscreen assistant) to 'Shopify Sidekick' (docked)
You can run it in Shadow DOM isolated mode so existing styles play nice
Everything has hooks and events, so you can add unique flavor (and easily share back!) in a few lines
We have a ton of demos on the library site, along with all the knobs to play with how each aspect renders. Tool and reasoning, custom loading animations, voice, approval UX, etc
We've also added a ton of examples across agent and frontend stacks
Check out the code @ https://github.com/runtypelabs/persona
Feedback and contributions welcome!"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Persona.js \u2013 a vanilla-JS agent UI library with native WebMCP (MIT)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/runtypelabs/persona"}},"_tags":["story","author_becomevocal","story_48612231","show_hn"],"author":"becomevocal","children":[48612298,48612372,48612429,48612433,48613011,48614321,48614524,48614831,48622263,48639673,48680427,48680887,48683532,48686849,48695832],"created_at":"2026-06-20T19:32:56Z","created_at_i":1781983976,"num_comments":19,"objectID":"48612231","points":27,"story_id":48612231,"story_text":"Hey everyone. My cofounder and I are open sourcing Persona.js (https://www.persona-chat.dev/). It's a VanillaJS library that helps anyone build agentic experiences on the web, without a framework dependency, and full WebMCP support
So, why'd we do this?
1) We're super fans of the web and the browser can do a ton today
2) We've seen AI builds be way overly complex because the FE requires a large project within an existing app OR the site wasn't using a framework to begin with
If you've been a part of huge apps with multiple frameworks inside them, or work inside CMS / website builders / ecomm platforms... you know what we're talking about. A "simple" AI feature disrupts your life for months
If you have a singular React codebase and can't possibly imagine building an interface without JSX, nothing to see here! You already have a lot of great options that are really cool too. Check out Assistant UI / CopilotKit / AI Elements which are all MIT.
There's a demo video here: https://www.youtube.com/watch?v=68D80uNsfH0.
Some specifics:
Persona has a ~15 kB brotli to first paint (the full widget lazy-loads on first click) while being able to render most of the primary agent experiences you see on the web: from 'Fin' (pill launcher) to 'Claude' (fullscreen assistant) to 'Shopify Sidekick' (docked)
You can run it in Shadow DOM isolated mode so existing styles play nice
Everything has hooks and events, so you can add unique flavor (and easily share back!) in a few lines
We have a ton of demos on the library site, along with all the knobs to play with how each aspect renders. Tool and reasoning, custom loading animations, voice, approval UX, etc
We've also added a ton of examples across agent and frontend stacks
Check out the code @ https://github.com/runtypelabs/persona
Feedback and contributions welcome!","title":"Show HN: Persona.js \u2013 a vanilla-JS agent UI library with native WebMCP (MIT)","updated_at":"2026-07-24T09:39:21Z","url":"https://github.com/runtypelabs/persona"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"swiftlyTyped"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"Open-source & web based MCP client. Chat with any MCP server in your own app. Powered by CopilotKit & Composio.
Shipped this with cursor over the weekend and built: \n1. The first web-based MCP client (deployed version on GH) \n2. An open-source client you can add into any app.
Built using CopilotKit for the client and interactivity layer + agent frontend which connects to a LangGraph ReAct agent that coordinates MCP calls. Uses Composio's MCP server."},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: Open MCP Client"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/CopilotKit/open-mcp-client"}},"_tags":["story","author_swiftlyTyped","story_43322812","show_hn"],"author":"swiftlyTyped","created_at":"2025-03-10T17:22:55Z","created_at_i":1741627375,"num_comments":0,"objectID":"43322812","points":6,"story_id":43322812,"story_text":"Open-source & web based MCP client. Chat with any MCP server in your own app. Powered by CopilotKit & Composio.
Shipped this with cursor over the weekend and built: \n1. The first web-based MCP client (deployed version on GH) \n2. An open-source client you can add into any app.
Built using CopilotKit for the client and interactivity layer + agent frontend which connects to a LangGraph ReAct agent that coordinates MCP calls. Uses Composio's MCP server.","title":"Show HN: Open MCP Client","updated_at":"2025-03-10T18:08:42Z","url":"https://github.com/CopilotKit/open-mcp-client"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"swiftlyTyped"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"`<CopilotTextarea />` is a drop-in `<textarea/>` replacement (looks identical & supports all customizations & params) w GPT completions (like in Gmail / GitHub Copilot).
It's built with slateJS and works with any OpenAI-compatible endpoint (bring your own backend).
You pass it a "purpose" prompt & external context to inform completions.\nE.g. the demo gif[1] has a purpose of "COOL and SLICK announcement post", and was given the CopilotKit release notes as context. (super important, to make suggestions specific).
Part of CopilotKit (open-source copilot/sidekick infra).
IMO it works really well. After working with it in all my apps for the past week it's honestly very frustrating to not have it available universally.
[1]: https://github.com/RecursivelyAI/CopilotKit"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: CopilotTextarea: W GPT Completions Like Gmail/GitHub. OSS"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"https://github.com/RecursivelyAI/CopilotKit"}},"_tags":["story","author_swiftlyTyped","story_37294797","show_hn"],"author":"swiftlyTyped","created_at":"2023-08-28T14:42:24Z","created_at_i":1693233744,"num_comments":0,"objectID":"37294797","points":5,"story_id":37294797,"story_text":"`<CopilotTextarea />` is a drop-in `<textarea/>` replacement (looks identical & supports all customizations & params) w GPT completions (like in Gmail / GitHub Copilot).
It's built with slateJS and works with any OpenAI-compatible endpoint (bring your own backend).
You pass it a "purpose" prompt & external context to inform completions.\nE.g. the demo gif[1] has a purpose of "COOL and SLICK announcement post", and was given the CopilotKit release notes as context. (super important, to make suggestions specific).
Part of CopilotKit (open-source copilot/sidekick infra).
IMO it works really well. After working with it in all my apps for the past week it's honestly very frustrating to not have it available universally.
[1]: https://github.com/RecursivelyAI/CopilotKit","title":"Show HN: CopilotTextarea: W GPT Completions Like Gmail/GitHub. OSS","updated_at":"2024-09-20T15:04:56Z","url":"https://github.com/RecursivelyAI/CopilotKit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ulidabess"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["copilotkit"],"value":"CopilotKit is an open-source typescript library giving you the ability to inject powerful AI-features anywhere users write text.
Current Features: \n1. Set Purpose Prompts\n2. AI auto-complete\n3.Refrence External Context \n4.Insertions"},"title":{"matchLevel":"none","matchedWords":[],"value":"Show HN: CopilotTextarea/> = an open-source, AI-infused react