Pica AI SDK

The Pica AI SDK is a TypeScript library for integrating Pica with Vercel's AI SDK.

Installation

npm install @picahq/ai

Setup

1. Create a new Pica account
2. Create a Connection via the Pica Dashboard
3. Create a Pica API key
4. Set the API key as an environment variable: PICA_SECRET_KEY=<your-api-key>

Configuration

The Pica SDK can be configured with the following options:

| Option | Type | Required | Default | Description | |--------|------|----------|---------|-------------| | serverUrl | String | No | https://api.picaos.com | URL for self-hosted Pica server | | connectors | String[] | No | - | List of connector keys to filter by. Pass ["*"] to initialize all available connectors, or specific connector keys to filter. If empty, no connections will be initialized | | identity | String | No | None | Filter connections by specific identifier | | identityType | "user" | "team" | "organization" | No | None | Filter connections by identity type | | authkit | Boolean | No | false | If true, the SDK will use Authkit to connect to prompt the user to connect to a platform that they do not currently have access to | | knowledgeAgent | Boolean | No | false | If true, the SDK will never execute actions, but will use Pica's knowledge to generate code. If true, use pica.intelligenceTool instead of pica.oneTool |

Usage

The Pica AI SDK is designed to work seamlessly with Vercel AI SDK. Here's an example implementation:

Express Example

1. Install dependencies

npm install express @ai-sdk/openai ai @picahq/ai dotenv

2. Create the server

import express from "express";
import { openai } from "@ai-sdk/openai";
import { generateText } from "ai";
import { Pica } from "@picahq/ai";
import * as dotenv from "dotenv";

dotenv.config();

const app = express();
const port = process.env.PORT || 3000;

app.use(express.json());

app.post("/api/ai", async (req, res) => {
  try {
    const { message } = req.body;

    // Initialize Pica
    const pica = new Pica(process.env.PICA_SECRET_KEY);

    // Generate the system prompt
    const systemPrompt = await pica.generateSystemPrompt(
      // Optional: Custom system prompt to append
    );

    // Create the stream
    const { text } = await generateText({
      model: openai("gpt-4o"),
      system: systemPrompt,
      tools: { ...pica.oneTool },
      prompt: message,
      maxSteps: 5,
    });

    res.setHeader("Content-Type", "application/json");
    
    res.status(200).json({ text });
  } catch (error) {
    console.error("Error processing AI request:", error);

    res.status(500).json({ error: "Internal server error" });
  }
});

app.listen(port, () => {
  console.log(`Server is running on port ${port}`);
});

export default app;

3. Test the server

curl --location 'http://localhost:3000/api/ai' \
--header 'Content-Type: application/json' \
--data '{
    "message": "What connections do I have access to?"
}'

Next.js Example

import { openai } from "@ai-sdk/openai";
import { convertToCoreMessages, streamText } from "ai";
import { Pica } from "@picahq/ai";

export async function POST(request: Request) {
  const { messages } = await request.json();

  const pica = new Pica(process.env.PICA_SECRET_KEY as string);

  const systemPrompt = await pica.generateSystemPrompt();

  const stream = streamText({
    model: openai("gpt-4o"),
    system: systemPrompt,
    tools: { ...pica.oneTool },
    messages: convertToCoreMessages(messages),
    maxSteps: 5,
  });

  return (await stream).toDataStreamResponse();
}

⭐️ You can see a full Next.js demo of the Pica AI SDK in action here