How to integrate Encodian MCP with Vercel AI SDK v6

Trusted by
AWS
Glean
Zoom
Airtable
GET STARTED FOR FREEGET A DEMO

30 min · no commitment · see it on your stack

Encodian logo
Vercel AI SDK logo
divider

Introduction

This guide walks you through connecting Encodian to Vercel AI SDK v6 using the Composio tool router. By the end, you'll have a working Encodian agent that can resize all images in project folder, extract author and page count from pdf, add custom header to every pdf file through natural language commands.

This guide will help you understand how to give your Vercel AI SDK agent real control over a Encodian account through Composio's Encodian MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

Also integrate Encodian with

TL;DR

Here's what you'll learn:
  • How to set up and configure a Vercel AI SDK agent with Encodian integration
  • Using Composio's Tool Router to dynamically load and access Encodian tools
  • Creating an MCP client connection using HTTP transport
  • Building an interactive CLI chat interface with conversation history management
  • Handling tool calls and results within the Vercel AI SDK framework

What is Vercel AI SDK?

The Vercel AI SDK is a TypeScript library for building AI-powered applications. It provides tools for creating agents that can use external services and maintain conversation state.

Key features include:

  • streamText: Core function for streaming responses with real-time tool support
  • MCP Client: Built-in support for Model Context Protocol via @ai-sdk/mcp
  • Step Counting: Control multi-step tool execution with stopWhen: stepCountIs()
  • OpenAI Provider: Native integration with OpenAI models

What is the Encodian MCP server, and what's possible with it?

The Encodian MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Encodian account. It provides structured and secure access to your Encodian document automation suite, so your agent can perform actions like managing files, processing PDFs, encoding or decoding content, and automating workflow tasks on your behalf.

  • File management and property retrieval: Quickly get detailed information about files, move documents between containers, and keep your Microsoft 365 storage organized automatically.
  • Document processing and automation: Direct your agent to add headers and footers to PDFs, extract PDF metadata, or resize images for seamless document formatting and compliance tasks.
  • Base64 content conversion: Effortlessly encode text or files to Base64, or decode Base64 strings back to usable files for secure data exchange and workflow integration.
  • Archive extraction and manipulation: Unzip files and retrieve their contents, making it easy to automate bulk document handling or trigger downstream processing steps.
  • Data integrity and comparison tools: Use hashing and text comparison utilities to verify file integrity, detect changes, or ensure consistency across your documents and workflows.

Supported Tools & Triggers

Tools
Add Attachments to PDFTool to add file attachments to a PDF document.
Add Image Watermark to PDFTool to add an image watermark to a PDF document.
Add Image Watermark to PDF (Advanced)Tool to add an advanced image watermark to a PDF with precise control over positioning, opacity, scale, quality, and rotation.
Array Add ItemsTool to add items to a JSON array at a specified position (first, last, or specific index).
Create ZIP ArchiveTool to create a ZIP archive from multiple documents.
Apply AI OCR to PDFTool to apply AI-powered OCR to a PDF document with optional preprocessing filters.
Apply OCR to PDF (Standard)Tool to apply standard OCR to a PDF document with optional preprocessing filters.
Decode Base64 StringTool to decode a Base64 string to a file.
Base64 EncodeTool to encode a string to Base64.
Calculate DateTool to calculate a date by adding or subtracting a time interval from a given date.
Check Array Contains ValueTool to check if a value exists within a JSON array.
Check Text Contains ValueTool to check if a text string contains a specific value with configurable comparison rules.
Clean StringTool to clean text by removing control characters, invalid filename characters, and custom character sets.
Clean Up Photo ImageTool to clean up photo images by removing artifacts, correcting orientation, and enhancing quality.
Combine ArraysTool to combine two JSON arrays by matching a key attribute.
Compare TextTool to compare two text strings and determine if they match.
Compare Word DocumentsTool to compare two Microsoft Word or PDF documents and generate a document with tracked changes.
Compress ImageTool to compress an image in JPG or PNG format.
Compress PDFTool to compress a PDF document by optimizing images, removing unused objects, and applying various compression techniques.
Concatenate TextTool to concatenate an array of text values with an optional delimiter.
Array to JSONTool to convert an array to a named JSON object.
Array to XMLTool to convert a JSON array to XML format.
Convert File to PDFTool to convert a file to PDF format.
Convert HTML to ImageTool to convert HTML content to an image.
Convert HTML to PDF (V2)Tool to convert HTML content or a URL to PDF format (V2).
Convert HTML to WordTool to convert HTML content to Word (DOCX) format.
Convert Image to GrayscaleTool to convert an image to grayscale.
Convert Image to PDFTool to convert an image file to PDF format with optional OCR.
Convert JSON to ExcelTool to convert JSON data to Excel format.
Convert JSON to XMLTool to convert JSON data to XML format.
Convert Time ZoneTool to convert a date and time value from one time zone to another using Encodian's time zone conversion API.
Convert XML to JSONTool to convert XML strings to JSON format.
Count Array ItemsTool to count the number of items in a JSON array or object.
Create QR CodeTool to generate a QR code barcode image with customizable size, colors, border, and encoding options.
Format Text CaseTool to format text with various case transformations (uppercase, lowercase, title case, etc.
Hash DataTool to compute a cryptographic hash (MD5, SHA256, etc.
Unzip FileExtracts all files from a ZIP archive and returns their base64-encoded contents.
Get Convert Excel SchemaTool to retrieve the dynamic schema for Excel conversion operations.
Get Convert Word SchemaTool to retrieve the dynamic JSON schema for Word document conversion operations.
Get Convert CAD SchemaTool to retrieve the dynamic schema for CAD file conversion operations.
Get Convert Image to PDF SchemaTool to retrieve the dynamic schema for Convert - Image to PDF operations.
Get Convert PowerPoint SchemaTool to retrieve the dynamic schema for PowerPoint conversion operations.
Get Convert Visio SchemaTool to retrieve the dynamic schema for Visio file conversion.
Get Create Barcode SchemaTool to retrieve the dynamic schema for creating a barcode.
Get Crop Image SchemaTool to retrieve the dynamic schema for the Crop Image action.
Get Dynamic Schema for HTTP RequestTool to retrieve the dynamic schema for the HTTP Request utility based on authentication type.
Get Word Insert Text SchemaTool to retrieve the dynamic schema for Word Insert Text operations.
Get File PropertiesTool to retrieve properties of a file.
Get Operation Status for AIRunPromptTextTool to get the operation status of an AIRunPromptText operation.
Get Operation Status for Encodian Send to FilerTool to get the operation status for an Encodian Send to Filer operation.
Get Operation Status Extract ImageTool to retrieve the operation status of a PDF ExtractImage operation.
Get Operation Status for ExtractTextRegionTool to retrieve the operation status of an ExtractTextRegion operation.
Get Operation Status File OnlyTool to retrieve operation status for file-only operations.
Get Operation Status for Image Extract TextTool to get the operation status of an ImageExtractText operation.
Get Operation Status for Multiple FilesTool to retrieve the operation status of a Word MultipleFiles operation.
Get Operation Status - PDF Split BarcodeTool to retrieve operation status for a PDF split barcode operation.
Get Operation Status for Split DocumentTool to retrieve the operation status of a PDF SplitDocument operation.
Get Sign PDF SchemaTool to retrieve the dynamic schema for PDF signing operations.
Get Subscription StatusTool to retrieve Encodian subscription status for Flowr and Vertr.
Resize ImageTool to resize an image by percentage or dimensions.
Move FileTool to move a file between containers.
Add PDF Header FooterTool to add HTML header and footer to a PDF.
Get PDF MetadataExtract comprehensive metadata and properties from PDF documents.
Watermark PDFTool to apply a text watermark to a PDF.
Read QR Code from DocumentTool to read QR codes from PDF or DOCX documents.
Word - Replace Text With ImageTool to replace text with an image in a Word document.
Validate Email AddressValidates an email address string against a custom regex pattern using Encodian's validation API.
Validate URL AvailabilityTool to validate the availability of a specified URL.
Write Range to ExcelTool to write values to a cell range in an Excel worksheet.

What is the Composio tool router, and how does it fit here?

What is Composio SDK?

Composio's Composio SDK helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Composio SDK

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Composio SDK works

The Composio SDK follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before you begin, make sure you have:
  • Node.js and npm installed
  • A Composio account with API key
  • An OpenAI API key

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key
  • Log in to the Composio dashboard.
  • Navigate to your API settings and generate a new API key.
  • Store this key securely as you'll need it for authentication.

Install required dependencies

bash
npm install @ai-sdk/openai @ai-sdk/mcp @composio/core ai dotenv

First, install the necessary packages for your project.

What you're installing:

  • @ai-sdk/openai: Vercel AI SDK's OpenAI provider
  • @ai-sdk/mcp: MCP client for Vercel AI SDK
  • @composio/core: Composio SDK for tool integration
  • ai: Core Vercel AI SDK
  • dotenv: Environment variable management

Set up environment variables

bash
OPENAI_API_KEY=your_openai_api_key_here
COMPOSIO_API_KEY=your_composio_api_key_here
COMPOSIO_USER_ID=your_user_id_here

Create a .env file in your project root.

What's needed:

  • OPENAI_API_KEY: Your OpenAI API key for GPT model access
  • COMPOSIO_API_KEY: Your Composio API key for tool access
  • COMPOSIO_USER_ID: A unique identifier for the user session

Import required modules and validate environment

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});
What's happening:
  • We're importing all necessary libraries including Vercel AI SDK's OpenAI provider and Composio
  • The dotenv/config import automatically loads environment variables
  • The MCP client import enables connection to Composio's tool server

Create Tool Router session and initialize MCP client

typescript
async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["encodian"],
  });

  const mcpUrl = session.mcp.url;
What's happening:
  • We're creating a Tool Router session that gives your agent access to Encodian tools
  • The create method takes the user ID and specifies which toolkits should be available
  • The returned mcp object contains the URL and authentication headers needed to connect to the MCP server
  • This session provides access to all Encodian-related tools through the MCP protocol

Connect to MCP server and retrieve tools

typescript
const mcpClient = await createMCPClient({
  transport: {
    type: "http",
    url: mcpUrl,
    headers: session.mcp.headers, // Authentication headers for the Composio MCP server
  },
});

const tools = await mcpClient.tools();
What's happening:
  • We're creating an MCP client that connects to our Composio Tool Router session via HTTP
  • The mcp.url provides the endpoint, and mcp.headers contains authentication credentials
  • The type: "http" is important - Composio requires HTTP transport
  • tools() retrieves all available Encodian tools that the agent can use

Initialize conversation and CLI interface

typescript
let messages: ModelMessage[] = [];

console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
console.log(
  "Ask any questions related to encodian, like summarize my last 5 emails, send an email, etc... :)))\n",
);

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
  prompt: "> ",
});

rl.prompt();
What's happening:
  • We initialize an empty messages array to maintain conversation history
  • A readline interface is created to accept user input from the command line
  • Instructions are displayed to guide the user on how to interact with the agent

Handle user input and stream responses with real-time tool feedback

typescript
rl.on("line", async (userInput: string) => {
  const trimmedInput = userInput.trim();

  if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
    console.log("\nGoodbye!");
    rl.close();
    process.exit(0);
  }

  if (!trimmedInput) {
    rl.prompt();
    return;
  }

  messages.push({ role: "user", content: trimmedInput });
  console.log("\nAgent is thinking...\n");

  try {
    const stream = streamText({
      model: openai("gpt-5"),
      messages,
      tools,
      toolChoice: "auto",
      stopWhen: stepCountIs(10),
      onStepFinish: (step) => {
        for (const toolCall of step.toolCalls) {
          console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\n👋 Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});
What's happening:
  • We use streamText instead of generateText to stream responses in real-time
  • toolChoice: "auto" allows the model to decide when to use Encodian tools
  • stopWhen: stepCountIs(10) allows up to 10 steps for complex multi-tool operations
  • onStepFinish callback displays which tools are being used in real-time
  • We iterate through the text stream to create a typewriter effect as the agent responds
  • The complete response is added to conversation history to maintain context
  • Errors are caught and displayed with helpful retry suggestions

Complete Code

Here's the complete code to get you started with Encodian and Vercel AI SDK:

typescript
import "dotenv/config";
import { openai } from "@ai-sdk/openai";
import { Composio } from "@composio/core";
import * as readline from "readline";
import { streamText, type ModelMessage, stepCountIs } from "ai";
import { createMCPClient } from "@ai-sdk/mcp";

const composioAPIKey = process.env.COMPOSIO_API_KEY;
const composioUserID = process.env.COMPOSIO_USER_ID;

if (!process.env.OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set");
if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set");

const composio = new Composio({
  apiKey: composioAPIKey,
});

async function main() {
  // Create a tool router session for the user
  const session = await composio.create(composioUserID!, {
    toolkits: ["encodian"],
  });

  const mcpUrl = session.mcp.url;

  const mcpClient = await createMCPClient({
    transport: {
      type: "http",
      url: mcpUrl,
      headers: session.mcp.headers, // Authentication headers for the Composio MCP server
    },
  });

  const tools = await mcpClient.tools();

  let messages: ModelMessage[] = [];

  console.log("Chat started! Type 'exit' or 'quit' to end the conversation.\n");
  console.log(
    "Ask any questions related to encodian, like summarize my last 5 emails, send an email, etc... :)))\n",
  );

  const rl = readline.createInterface({
    input: process.stdin,
    output: process.stdout,
    prompt: "> ",
  });

  rl.prompt();

  rl.on("line", async (userInput: string) => {
    const trimmedInput = userInput.trim();

    if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) {
      console.log("\nGoodbye!");
      rl.close();
      process.exit(0);
    }

    if (!trimmedInput) {
      rl.prompt();
      return;
    }

    messages.push({ role: "user", content: trimmedInput });
    console.log("\nAgent is thinking...\n");

    try {
      const stream = streamText({
        model: openai("gpt-5"),
        messages,
        tools,
        toolChoice: "auto",
        stopWhen: stepCountIs(10),
        onStepFinish: (step) => {
          for (const toolCall of step.toolCalls) {
            console.log(`[Using tool: ${toolCall.toolName}]`);
          }
          if (step.toolCalls.length > 0) {
            console.log(""); // Add space after tool calls
          }
        },
      });

      for await (const chunk of stream.textStream) {
        process.stdout.write(chunk);
      }

      console.log("\n\n---\n");

      // Get final result for message history
      const response = await stream.response;
      if (response?.messages?.length) {
        messages.push(...response.messages);
      }
    } catch (error) {
      console.error("\nAn error occurred while talking to the agent:");
      console.error(error);
      console.log(
        "\nYou can try again or restart the app if it keeps happening.\n",
      );
    } finally {
      rl.prompt();
    }
  });

  rl.on("close", async () => {
    await mcpClient.close();
    console.log("\n👋 Session ended.");
    process.exit(0);
  });
}

main().catch((err) => {
  console.error("Fatal error:", err);
  process.exit(1);
});

Conclusion

You've successfully built a Encodian agent using the Vercel AI SDK with streaming capabilities! This implementation provides a powerful foundation for building AI applications with natural language interfaces and real-time feedback.

Key features of this implementation:

  • Real-time streaming responses for a better user experience with typewriter effect
  • Live tool execution feedback showing which tools are being used as the agent works
  • Dynamic tool loading through Composio's Tool Router with secure authentication
  • Multi-step tool execution with configurable step limits (up to 10 steps)
  • Comprehensive error handling for robust agent execution
  • Conversation history maintenance for context-aware responses

You can extend this further by adding custom error handling, implementing specific business logic, or integrating additional Composio toolkits to create multi-app workflows.

How to build Encodian MCP Agent with another framework

OpenAI Agents SDK logo

OpenAI Agents SDK

Use Encodian MCP with OpenAI Agents SDK

Claude Agent SDK logo

Claude Agent SDK

Use Encodian MCP with Claude Agent SDK

Claude Code logo

Claude Code

Use Encodian MCP with Claude Code

Claude Cowork logo

Claude Cowork

Use Encodian MCP with Claude Cowork

Codex logo

Codex

Use Encodian MCP with Codex

OpenClaw logo

OpenClaw

Use Encodian MCP with OpenClaw

Hermes logo

Hermes

Use Encodian MCP with Hermes

CLI logo

CLI

Use Encodian MCP with CLI

Google ADK logo

Google ADK

Use Encodian MCP with Google ADK

LangChain logo

LangChain

Use Encodian MCP with LangChain

Mastra AI logo

Mastra AI

Use Encodian MCP with Mastra AI

LlamaIndex logo

LlamaIndex

Use Encodian MCP with LlamaIndex

CrewAI logo

CrewAI

Use Encodian MCP with CrewAI

FAQ

What are the differences in Tool Router MCP and Encodian MCP?

With a standalone Encodian MCP server, the agents and LLMs can only access a fixed set of Encodian tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Encodian and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with Vercel AI SDK v6?

Yes, you can. Vercel AI SDK v6 fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Encodian tools.

Can I manage the permissions and scopes for Encodian while using Tool Router?

Yes, absolutely. You can configure which Encodian scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Encodian data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.

Get started for freeGet a demo↗