Writing agent state
Write to agent's state from your application.
This example demonstrates writing to shared state in the CopilotKit Feature Viewer.
What is this?
This guide shows you how to write to your agent's state from your application.
When should I use this?
You can use this when you want to provide the user with feedback about what your agent is doing, specifically when your agent is calling tools. CopilotKit allows you to fully customize how these tools are rendered in the chat.
Implementation
Run and connect your agent
You'll need to run your agent and connect it to CopilotKit before proceeding.
If you don't already have CopilotKit and your agent connected, choose one of the following options:
Define the Agent State
Create your Pydantic AI agent with a stateful structure. Here's a complete example that tracks language:
from textwrap import dedent
from pydantic import BaseModel
from pydantic_ai import Agent, RunContext
from pydantic_ai.ag_ui import StateDeps
class AgentState(BaseModel):
"""State for the agent."""
language: str = "english"
agent = Agent("openai:gpt-4o-mini", deps_type=StateDeps[AgentState])
@agent.instructions()
async def language_instructions(ctx: RunContext[StateDeps[AgentState]]) -> str:
"""Instructions for the language tracking agent.
Args:
ctx: The run context containing language state information.
Returns:
Instructions string for the language tracking agent.
"""
return dedent(
f"""
You are a helpful assistant for tracking the language.
IMPORTANT:
- ALWAYS use the lower case for the language
- ALWAYS response in the current language: {ctx.deps.state.language}
"""
)
app = agent.to_ag_ui(deps=StateDeps(AgentState()))
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=8000)Call setState function from the useCoAgent hook
useCoAgent returns a setState function that you can use to update the agent state. Calling this
will update the agent state and trigger a rerender of anything that depends on the agent state.
import { useCoAgent } from "@copilotkit/react-core";
// Define the agent state type, should match the actual state of your agent
type AgentState = {
language: "english" | "spanish";
}
// Example usage in a pseudo React component
function YourMainContent() {
const { state, setState } = useCoAgent<AgentState>({
name: "my_agent", // MUST match the agent name in CopilotRuntime
initialState: { language: "english" } // optionally provide an initial state
});
// ...
const toggleLanguage = () => {
setState({ language: state.language === "english" ? "spanish" : "english" });
};
// ...
return (
// style excluded for brevity
<div>
<h1>Your main content</h1>
<p>Language: {state.language}</p>
<button onClick={toggleLanguage}>Toggle Language</button>
</div>
);
}Important
The name parameter must exactly match the agent name you defined in your CopilotRuntime configuration (e.g., my_agent from the quickstart).
Give it a try!
You can now use the setState function to update the agent state and state to read it. Try toggling the language button
and talking to your agent. You'll see the language change to match the agent's state.
Advanced Usage
Re-run the agent with a hint about what's changed
The new agent state will be used next time the agent runs.
If you want to re-run it manually, use the run argument on the useCoAgent hook.
The agent will be re-run, and it will get not only the latest updated state, but also a hint that can depend on the data delta between the previous and the current state.
import { useCoAgent } from "@copilotkit/react-core";
import { TextMessage, MessageRole } from "@copilotkit/runtime-client-gql";
// ...
function YourMainContent() {
const { state, setState, run } = useCoAgent<AgentState>({
name: "my_agent", // MUST match the agent name in CopilotRuntime
initialState: { language: "english" } // optionally provide an initial state
});
// setup to be called when some event in the app occurs
const toggleLanguage = () => {
const newLanguage = state.language === "english" ? "spanish" : "english";
setState({ language: newLanguage });
// re-run the agent and provide a hint about what's changed
run(({ previousState, currentState }) => {
return new TextMessage({
role: MessageRole.User,
content: `the language has been updated to ${currentState.language}`,
});
});
};
return (
// ...
);
}Intermediately Stream and Render Agent State
By default, the Pydantic AI Agent state will only update between Pydantic AI Agent node transitions -- which means state updates will be discontinuous and delayed.
