Llama Cloud Mcp

NPM version

The official TypeScript SDK for LlamaParse - the enterprise platform for agentic OCR and document processing.

With this SDK, create powerful workflows across many features:

• Parse - Agentic OCR and parsing for 130+ formats
• Extract - Structured data extraction with custom schemas
• Classify - Document categorization with natural-language rules
• Agents - Deploy document agents as APIs
• Index - Document ingestion and embedding for RAG

Documentation

• Get an API Key
• Getting Started Guide
• Full API Reference

Installation

npm install @llamaindex/llama-cloud

Quick Start

import LlamaCloud from '@llamaindex/llama-cloud';

const client = new LlamaCloud({
  apiKey: process.env['LLAMA_CLOUD_API_KEY'], // This is the default and can be omitted
});

// Parse a document
const job = await client.parsing.create({
  tier: 'agentic',
  version: 'latest',
  file_id: 'your-file-id',
});

console.log(job.id);

File Uploads

import fs from 'fs';
import LlamaCloud from '@llamaindex/llama-cloud';

const client = new LlamaCloud();

// Upload using a file stream
await client.files.create({
  file: fs.createReadStream('/path/to/document.pdf'),
  purpose: 'purpose',
});

// Or using a File object
await client.files.create({
  file: new File(['content'], 'document.txt'),
  purpose: 'purpose',
});

MCP Server

Use the Llama Cloud MCP Server to enable AI assistants to interact with the API:

Add to Cursor Install in VS Code

Error Handling

When the API returns a non-success status code, an APIError subclass is thrown:

await client.pipelines.list({ project_id: 'my-project-id' }).catch((err) => {
  if (err instanceof LlamaCloud.APIError) {
    console.log(err.status); // 400
    console.log(err.name); // BadRequestError
  }
});

Retries and Timeouts

The SDK automatically retries requests 2 times on connection errors, timeouts, rate limits, and 5xx errors. Requests timeout after 1 minute by default. Functions that combine multiple API calls (e.g. client.parsing.parse()) will have larger timeouts by default to account for the multiple requests and polling.

const client = new LlamaCloud({
  maxRetries: 0, // Disable retries (default: 2)
  timeout: 30 * 1000, // 30 second timeout (default: 1 minute)
});

Pagination

List methods support auto-pagination with for await...of:

async function fetchAllExtractV2Jobs(params) {
  const allExtractV2Jobs = [];
  // Automatically fetches more pages as needed.
  for await (const extractV2Job of client.extract.list({ page_size: 20 })) {
    allExtractV2Jobs.push(extractV2Job);
  }
  return allExtractV2Jobs;
}

Or fetch one page at a time:

let page = await client.extract.list({ page_size: 20 });
for (const extractV2Job of page.items) {
  console.log(extractV2Job);
}

while (page.hasNextPage()) {
  page = await page.getNextPage();
}

Logging

Configure logging via the LLAMA_CLOUD_LOG environment variable or the logLevel option:

const client = new LlamaCloud({
  logLevel: 'debug', // 'debug' | 'info' | 'warn' | 'error' | 'off'
});

Requirements

• TypeScript >= 4.9
• Node.js 20+, Deno 1.28+, Bun 1.0+, or modern browsers

Contributing

See CONTRIBUTING.md.