Run Pipe Agent Chat with LLM API Keys
This example demonstrates how to run a pipe agent chat with LLM API keys.
Run Pipe Agent Chat with LLM API Keys Example
Run Pipe Agent Chat with LLM API Keys Example
import 'dotenv/config';
import {getRunner, Langbase} from 'langbase';
const langbase = new Langbase({
apiKey: process.env.LANGBASE_API_KEY!
});
async function main() {
await createSummaryAgent();
// Get readable stream
const {stream, threadId, rawResponse} = await langbase.pipes.run({
stream: true,
name: 'summary-agent',
rawResponse: true,
messages: [
{
role: 'user',
content: 'Who is an AI Engineer?'
}
],
llmKey: process.env.LLM_KEY!, // Your LLM API key
});
// Convert the stream to a stream runner.
const runner = getRunner(stream);
// Method 1: Using event listeners
runner.on('connect', () => {
console.log('Stream started.\n');
});
runner.on('content', content => {
process.stdout.write(content);
});
runner.on('end', () => {
console.log('\nStream ended.');
});
runner.on('error', error => {
console.error('Error:', error);
});
}
/**
* Creates a summary agent pipe if it doesn't already exist.
*
* This function checks if a pipe with the name 'summary-agent' exists in the system.
* If the pipe doesn't exist, it creates a new private pipe with a system message
* configuring it as a helpful assistant.
*
* @async
* @returns {Promise<void>} A promise that resolves when the operation is complete
* @throws {Error} Logs any errors encountered during the creation process
*/
async function createSummaryAgent() {
try {
await langbase.pipes.create({
name: 'summary-agent',
upsert: true,
status: 'private',
messages: [
{
role: 'system',
content: 'You are a helpful assistant that help users summarize text.',
},
],
});
} catch (error) {
console.error('Error creating summary agent:', error);
}
}
main();