useAllemAgent

Agent hook with status tracking, step history, and tool call awareness. Built on the Vercel AI SDK.

import { useAllemAgent } from "@allem-sdk/agents";

Options

Prop	Type	Default	Description
api	string	—	Override API endpoint from AllemAIProvider
model	string	—	Model identifier (e.g. "gemini-2.0-flash")
provider	"google" \| "anthropic" \| "openai"	—	Provider name
systemPrompt	string	—	System prompt sent with requests
headers	Record<string, string>	—	Extra headers merged with provider headers
experimental_throttle	number	—	Throttle wait in ms for updates

Returns

All properties from UseChatHelpers, plus:

Property	Type	Description
agentStatus	AgentStatus	"idle" \| "thinking" \| "calling-tool" \| "done"
steps	AgentStep[]	Ordered list of steps taken in the current turn
currentToolCalls	AgentToolCall[]	Tool calls from the most recent step
isAgentRunning	boolean	Whether the agent is actively processing
messages	UIMessage[]	Array of chat messages
sendMessage	({ text }): Promise	Send a message to the agent
status	"ready" \| "submitted" \| "streaming"	Underlying chat status
stop	() => void	Stop the agent

Usage

import { AllemAIProvider } from "@allem-sdk/ai";
import { useAllemAgent } from "@allem-sdk/agents";

function App() {
  return (
    <AllemAIProvider api="/api/agent" provider="google">
      <Agent />
    </AllemAIProvider>
  );
}

function Agent() {
  const {
    messages,
    sendMessage,
    agentStatus,
    steps,
    currentToolCalls,
    isAgentRunning,
  } = useAllemAgent();

  return (
    <div>
      {agentStatus === "calling-tool" && (
        <p>Using tool: {currentToolCalls[0]?.toolName}</p>
      )}
      {messages.map((m) => (
        <div key={m.id}>
          {m.parts?.filter((p) => p.type === "text").map((p) => p.text).join("")}
        </div>
      ))}
      <p>Steps taken: {steps.length}</p>
    </div>
  );
}