# Frontend Tools

> Let your agent interact with and update your application's UI.

> **Not available for Langchain DeepAgents yet.** The code on this page comes from the showcase's `frontend-tools` demo, and Langchain DeepAgents doesn't have it yet. See this page for [CopilotKit's Built-in Agent](/frontend-tools), [LangGraph (Python)](/langgraph-python/frontend-tools), [LangGraph (TypeScript)](/langgraph-typescript/frontend-tools).

<Callout type="info" title="See this in Inspector">
  Open Inspector on localhost. Go to **Inspect**, then **Frontend Tools**.
  Your tool and its schema are listed.

  More detail: [Inspector](/deepagents/inspector).
</Callout>

## What is this? [#what-is-this]

A frontend tool is a function that you register in the browser. Your agent can call it, and the function runs in the user's browser. Because the handler runs in the browser, it has direct access to component state, browser APIs, and any third-party UI library the page already uses. That's how an agent can "reach into" the app: update React state, trigger animations, read `localStorage`, pop a toast, or steer the user's view.

This page covers the "agent drives the UI" shape of frontend tools. To render a component for a tool call, see [Generative UI](/deepagents/generative-ui). To ask the user before a tool runs, see [Human-in-the-loop](/deepagents/human-in-the-loop).

## When should I use this? [#when-should-i-use-this]

Use frontend tools when your agent needs to:

- Read or modify React component state
- Access browser APIs like `localStorage`, `sessionStorage`, or cookies
- Trigger UI updates, animations, or transitions
- Show alerts, toasts, or notifications
- Interact with third-party frontend libraries

## How it works in code [#how-it-works-in-code]

<Steps>
  <Step>
    ### Wire CopilotKit middleware into your agent

    Deep Agents compile to LangGraph graphs, and `CopilotKitMiddleware` is the
    bridge. It reads the frontend tools from the request. It adds them to the
    tool list that the model sees. Without the middleware, a tool that you
    register with `useFrontendTool` or `useComponent` never reaches the model.

    <Tabs groupId="agent_language" items={['Python', 'TypeScript']} persist>
      <Tab value="Python">
        ```python title="agent.py"
        from deepagents import create_deep_agent
        from copilotkit import CopilotKitMiddleware

        agent = create_deep_agent(
            model="openai:gpt-6-luna",
            tools=[],  # Backend tools go here
            middleware=[CopilotKitMiddleware()],
            system_prompt=SYSTEM_PROMPT,
        )
        ```
      </Tab>
      <Tab value="TypeScript">
        In TypeScript, define your custom state as a middleware with
        `createMiddleware`, and pass it to `createDeepAgent` together with
        `copilotkitMiddleware`:

        ```ts title="agent.ts"
        import { createMiddleware } from "langchain";
        import { copilotkitMiddleware } from "@copilotkit/sdk-js/langgraph";
        import { z } from "zod";

        export const yourStateMiddleware = createMiddleware({
          name: "YourAgentState",
          stateSchema: z.object({
            yourAdditionalProperty: z.string().optional(),
          }),
        });

        // Pass both middlewares when constructing the agent:
        // createDeepAgent({ middleware: [yourStateMiddleware, copilotkitMiddleware], ... })
        ```
      </Tab>
    </Tabs>

    Without the middleware the run still completes, and the tool is never
    called, because the model was never told the tool exists. You do not
    write code to call a frontend tool: the middleware adds it to every model
    call, and the agent calls it like any other tool.

  </Step>
  <Step>
    ### Tell the model when to call it

    The forwarded tool arrives on every run, but a model with no instruction
    about it will answer in prose and never call it. Name the tool in
    `system_prompt` and say what it is for.

    ```python title="agent.py"
    SYSTEM_PROMPT = """
    You are a data visualization assistant.

    When the user asks for a chart, call `render_bar_chart` with a concise
    title and a `data` array of `{label, value}` items.

    Keep chat responses brief and let the chart do the talking.
    """
    ```

  </Step>
  <Step>
    ### Register a tool in the browser

    Deep Agents has no live demo on these pages yet, so the demo code below
    does not show for it. Here is a minimal frontend tool to start from. It
    says hello to the user:

    ```tsx title="page.tsx"
    import { z } from "zod";
    import { useFrontendTool } from "@copilotkit/react-core/v2";

    export function Page() {
      // ...

      useFrontendTool({
        name: "sayHello",
        description: "Say hello to the user",
        parameters: z.object({
          name: z.string().describe("The name of the user to say hello to"),
        }),
        handler: async ({ name }) => {
          alert(`Hello, ${name}!`);
          return `Said hello to ${name}!`;
        },
      });

      // ...
    }
    ```

    Ask the agent: `Say hello to Ada.` The browser shows an alert with
    `Hello, Ada!`, and the agent replies after it gets the result.

  </Step>
</Steps>

Register a frontend tool with `useFrontendTool`. Give it a name, a Zod schema for parameters, and a handler. The agent can then call it like any other tool and your frontend runs it in the browser.

The handler receives the parsed, type-safe parameters and can do anything
the browser can: update state, call an API, touch the DOM. Its return value
is sent back to the agent as the tool result so the model can reason about
what happened.

Ask the agent to change the background to a blue gradient. The page background changes, and the agent gets `{ status: "success" }` as the tool result.

## Registering a list of tools [#registering-a-list-of-tools]

`useFrontendTool` registers one tool per call, so it cannot be called in a loop
over a list whose length changes between renders. When the set of tools comes
from state, from props, or from a backend response, use
[`useFrontendTools`](/reference/hooks/useFrontendTools) instead. It takes an
array and runs a single effect over it, so the array can be empty on one render
and hold twenty entries on the next.

```tsx
useFrontendTools(
  reports.map((report) => ({
    name: `open_${report.id}`,
    description: `Open the ${report.title} report`,
    handler: async () => navigate(`/reports/${report.id}`),
  })),
  [navigate],
);
```

On each render:

- A tool that leaves the array is unregistered.
- A tool that joins the array is registered.
- A re-render that produces an equal list does not re-register anything.

A description built from your data stays current on its own. The second argument
is for values a handler closes over, such as `navigate` above.

<IntegrationGrid path="frontend-tools" />
