inkeep · nick-inkeep · Oct 12, 2025 · Oct 12, 2025 · Oct 12, 2025 · Oct 13, 2025
diff --git a/agents-docs/content/docs/concepts.mdx b/agents-docs/content/docs/concepts.mdx
@@ -1,13 +1,13 @@
 ---
 title: Core concepts
 sidebarTitle: Concepts
-description: Learn about the key building blocks of Inkeep - Sub Agents, Agents, tools, data components, and more.
+description: Learn about the key building blocks of Inkeep - Agents, Sub Agents, tools, data components, and more.
 icon: "LuBoxes"
 ---
 
 ## Agents
 
-In Inkeep, an Agent is the top-level entity you as an end-user can interface with via conversational experiences (chat) or trigger programmatically (via API).
+In Inkeep, an Agent is the top-level entity you interface with via conversational experiences (chat) or trigger programmatically (via API).
 
 Under the hood, an Agent is made up of one or more **Sub Agents** that work together to respond to a user or complete a task.
 
@@ -22,7 +22,7 @@ In Inkeep, tools can be added to Sub Agents as:
 - **MCP Servers**: A common way to connect to external services and APIs. Many SaaS providers provide out-of-the-box MCP Servers, but you can also create your own and register them with their associated **Credentials** on Inkeep for Agents to use.
 - **Function Tools**: Custom JavaScript functions that Agents can execute directly without the need for standing up an MCP server. 
 
-Typically, you want a Sub Agent to handle narrow, well-defined tasks. As a general rule of thumb, only add up to 5-7 tools to a Sub Agent.
+Typically, you want a Sub Agent to handle narrow, well-defined tasks. As a general rule of thumb, keep Sub Agents to be using 5-7 related tools at a time.
 
 ## Sub Agent relationships
 

diff --git a/agents-docs/content/docs/overview.mdx b/agents-docs/content/docs/overview.mdx
@@ -41,13 +41,13 @@ A code-first approach for building and managing teams of Agents with a multi-age
      id: "hello-agent",
      name: "Hello Agent",
      description: "Says hello",
-     prompt: 'You are a basic Agent that just says hello. You only reply with the word "hello", but you may do it in different variations like h3110, h3110w0rld, h3110w0rld! etc...',
+     prompt: 'Only reply with the word "hello", but you may do it in different variations like h3110, h3110w0rld, h3110w0rld! etc...',
    });
 
    export const basicAgent = agent({
      id: "basic-agent",
      name: "Basic Agent Example",
-     description: "A basic agent",
+     description: "A basic agent that just says hello",
      defaultSubAgent: helloAgent,
      subAgents: () => [helloAgent],
    });
@@ -58,10 +58,10 @@ The **Visual Builder and TypeScript SDK are fully interoperable**: your technica
 ## Platform Overview
 
 **Inkeep Open Source** includes:
-- A Visual Builder & TypeScript SDK with 2-way sync
+- A Visual Builder & TypeScript SDK with full 2-way sync
 - Multi-Agent architecture for building teams of agents
 - MCP Tools with Credential Management
-- A UI Component Library for dynamic AI chat experiences
+- A UI Component Library for rich AI chat experiences
 - Triggering Agents with MCP, A2A, & Vercel SDK APIs
 - Observability via a Traces UI & OpenTelemetry
 - Easy deployment to Vercel and Docker
@@ -78,7 +78,7 @@ Join our [community](/community/inkeep-community) to stay up to date and get sup
 
 <Cards>
   <Card title="Follow the Quick Start" icon="LuZap" href="/quick-start">
-    Get started with the Visual Builder and SDK in under 5 minutes.
+    Get started with the Visual Builder and TypeScript SDK in under 5 minutes.
   </Card>
   <Card title="Conceptual Overview" icon="LuLightbulb" href="/concepts">
     Learn about the key concepts of building Agents with Inkeep.

diff --git a/agents-docs/content/docs/quick-start/traces.mdx b/agents-docs/content/docs/quick-start/traces.mdx
@@ -1,5 +1,5 @@
 ---
-title: Traces and Live Debugger
+title: Live Debugger, Traces, and OTEL Telemetry 
 sidebarTitle: Traces
 description: Set up SigNoz to enable full observability with traces and live debugging capabilities for your agents.
 icon: "LuActivity"

diff --git a/agents-docs/content/docs/talk-to-your-agents/vercel-ai-sdk.mdx b/agents-docs/content/docs/talk-to-your-agents/vercel-ai-sdk.mdx
@@ -8,13 +8,48 @@ keywords: Vercel AI SDK, useChat, React, streaming, chat interface, agents
 
 ## Overview
 
-The Vercel AI SDK v5's `useChat` hook provides a React-friendly way to build chat interfaces that stream responses from your agent. This guide shows you how to configure `useChat` to connect to the agent framework's chat endpoint.
+Agents built with Inkeep have an `/api/chat` [API endpoint](/talk-to-your-agents/api) that is compatible with Vercel AI SDK's streaming protocol, so it works out of the box with Vercel's AI SDK ecosystem.
 
-<Tip>
-  The agent framework's `/api/chat` endpoint is compatible with Vercel AI SDK's streaming protocol. See [Via API](/talk-to-your-agents/api) for low-level HTTP details.
-</Tip>
+To build totally custom UIs, you have two options in the Vercel ecosystem:
+- Use the [AI Elements](https://ai-sdk.dev/elements/overview) component library for a shadcn-like library of Chat primitives.
+- Leverage `useChat` and other utilities to build custom UIs in React, Svelte, Vue.js, and Angular.
 
-## Installation
+## Using AI Elements
+
+For a production-ready UI components, consider using [Vercel AI Elements](https://ai-sdk.dev/elements/overview):
+
+```tsx
+import { Conversation, Message, PromptInput, Actions } from '@ai-sdk/react-elements';
+import { useChat } from '@ai-sdk/react';
+
+export default function ChatWithElements() {
+  const { messages, sendMessage } = useChat({
+    transport: new DefaultChatTransport({
+      api: "${process.env.NEXT_PUBLIC_INKEEP_AGENTS_RUN_API_URL}/api/chat", // 
+      headers: {
+        "x-inkeep-tenant-id": "default",
+        "x-inkeep-project-id": "weather-project",
+        "x-inkeep-agent-id": "weather-agent",
+      },
+    })
+  });
+
+  return (
+    <Conversation messages={messages}>
+      {messages.map(message => (
+        <Message key={message.id} message={message}>
+          <Actions message={message} />
+        </Message>
+      ))}
+      <PromptInput onSubmit={sendMessage} />
+    </Conversation>
+  );
+}
+```
+
+AI Elements includes components for reasoning panels, code blocks, tool execution displays, and more—all styled with shadcn/ui patterns.
+
+## Using `useChat`
 
 Install the Vercel AI SDK in your React application:
 
@@ -36,7 +71,7 @@ import { useState } from "react";
 export default function Page() {
   const { messages, sendMessage } = useChat({
     transport: new DefaultChatTransport({
-      api: "http://localhost:3003/api/chat",
+      api: "${process.env.NEXT_PUBLIC_INKEEP_AGENTS_RUN_API_URL}/api/chat", // 
       headers: {
         "x-inkeep-tenant-id": "default",
         "x-inkeep-project-id": "weather-project",
@@ -162,7 +197,7 @@ const { messages, sendMessage } = useChat({
 
 ## Complete Example
 
-Here's a more complete example with conversation management and styling:
+Here's a more complete example with conversation management and custom styling:
 
 ```tsx
 "use client";
@@ -174,7 +209,7 @@ import { useState } from "react";
 export default function Page() {
   const { messages, sendMessage } = useChat({
     transport: new DefaultChatTransport({
-      api: "http://localhost:3003/api/chat",
+      api: "{process.env.NEXT_PUBLIC_INKEEP_AGENTS_RUN_API_URL}/api/chat",
       headers: {
         "x-inkeep-tenant-id": "default",
         "x-inkeep-project-id": "weather-project",
@@ -267,7 +302,7 @@ export default function Page() {
 For production deployments, store configuration in environment variables:
 
 ```bash
-NEXT_PUBLIC_AGENTS_RUN_API_URL=http://localhost:3003
+NEXT_PUBLIC_AGENTS_RUN_API_URL=your_domain_for_run_api
 NEXT_PUBLIC_INKEEP_API_KEY=your-api-key
 NEXT_PUBLIC_TENANT_ID=your-tenant-id
 NEXT_PUBLIC_PROJECT_ID=your-project-id
@@ -281,6 +316,9 @@ NEXT_PUBLIC_AGENT_ID=your-agent-id
 ## Next Steps
 
 <Cards>
+  <Card title="AI Elements" icon="LuPackage" href="https://ai-sdk.dev/elements/overview">
+    Explore Vercel's prebuilt UI components library built on shadcn/ui for rapid development.
+  </Card>
   <Card title="React UI Components" icon="LuLayers" href="/talk-to-your-agents/react/chat-button">
     Use prebuilt React components for a complete chat UI experience.
   </Card>

diff --git a/agents-docs/content/docs/troubleshooting.mdx b/agents-docs/content/docs/troubleshooting.mdx
@@ -8,7 +8,7 @@ keywords: troubleshooting, debugging, errors, timeline, signoz, widget implement
 
 ## Overview
 
-When something breaks in your Inkeep agent system, follow this systematic approach to identify and resolve the issue. This guide provides a structured methodology for debugging problems across different components of your agent infrastructure.
+This guide provides a structured methodology for debugging problems across different components of your agent system.
 
 ## Step 1: Check the Timeline
 
@@ -31,7 +31,6 @@ Clicking on this error card reveals:
 
 This detailed error information helps you pinpoint exactly what went wrong and where in your agent's execution chain.
 
-
 ## Step 2: Check Signoz
 
 Signoz provides distributed tracing and observability for your agent system, offering deeper insights when the built-in timeline isn't sufficient.

diff --git a/agents-docs/content/docs/typescript-sdk/project-structure.mdx b/agents-docs/content/docs/typescript-sdk/project-structure.mdx
@@ -6,7 +6,7 @@ icon: "LuFolder"
 
 ## Overview
 
-Inkeep Agent projects follow a standardized directory structure that enables the CLI to automatically discover and manage your Sub Agents, agents, tools, and configurations. This convention-based approach simplifies project organization and deployment workflows.
+Inkeep Agent projects follow a standardized directory structure that enables the CLI to automatically discover and manage your Agents, Sub Agents, tools, and configurations. This convention-based approach simplifies project organization and deployment workflows.
 
 ## Standard Project Layout
 

diff --git a/agents-docs/content/docs/typescript-sdk/structured-outputs/data-components.mdx b/agents-docs/content/docs/typescript-sdk/structured-outputs/data-components.mdx
@@ -1,6 +1,7 @@
 ---
-title: Data Components
-description: Learn how to create and use data components for rich agent responses
+title: Create Rich UI Blocks with Data Components
+sidebarTitle: Custom UI Blocks
+description: Learn how to create and use data components for rich, interactive Agent messages.
 icon: "LuBlocks"
 ---
 

diff --git a/agents-docs/content/docs/visual-builder/structured-outputs/data-components.mdx b/agents-docs/content/docs/visual-builder/structured-outputs/data-components.mdx
@@ -1,6 +1,7 @@
 ---
-title: Data Components
-description: Data components are used to add data to the Agent.
+title: Create Rich UI Blocks with Data Components
+sidebarTitle: Custom UI Blocks
+description: Learn how to create and use data components for rich, interactive Agent messages.
 icon: "LuBlocks"
 ---