ToolJet

This guide provides a streamlined process for integrating Portkey AI into ToolJet using the ToolJet Marketplace Plugin. Follow this guide to add AI-powered capabilities such as chat completions and automations into your ToolJet apps, helping developers create powerful, low-code applications backed by Portkey’s AI for chat-based interfaces and automated workflows.

Prerequisites

Portkey API Key and Virtual Key from your Portkey dashboard.
ToolJet Account with access to the Marketplace Plugin feature.
Basic familiarity with ToolJet UI components.

Watch Demo

Watch this demo for a quick walkthrough on ToolJet’s UI components.

Before following this guide, ensure that you have completed the setup for using marketplace plugins in ToolJet.

Step-by-Step Onboarding

Install the Portkey Plugin from ToolJet Marketplace

Go to ToolJet Dashboard > Plugins > Marketplace.
Search for Portkey Plugin.
Click Install to add the plugin to your project.

If Portkey has already been added to your marketplace, you can skip this step.

Configure the Portkey Plugin

Navigate to Data Sources in your ToolJet workspace.
Open Plugins.
Click Add Portkey.

Add Portkey as a Datasource

Enter the following details:

Authorization: Your Portkey API Key
Default Virtual Key: Your Portkey Virtual Key

Test the connection to ensure the keys are configured correctly.

Use Portkey in Your ToolJet App

Go to Queries > Add Datasource > Select Portkey Plugin.
From the dropdown, Select an Operation (details in the next section).
Run the query to verify it responds correctly.

Link the Query to UI Components (Example)

Add Text Input and Button widgets to your app’s interface.
Configure the Button’s onClick action to Execute Query using the Portkey API.
Use a Text Box widget to display the query results.

Test and Deploy

Use Preview Mode to test the interaction between your app’s UI and the Portkey API.
Deploy the app from the ToolJet dashboard.

Supported Operations

Portkey supports the following operations within ToolJet:

Completion
Chat
Prompt Completion
Create Embedding

Completion

Generates text completions based on a given prompt.Parameters:

Prompt: Input text to generate completions for.
Model: The AI model to use.
Max Tokens: Maximum number of tokens to generate.
Temperature: Controls randomness.
Stop Sequences: Sequences where the API stops generating further tokens.
Metadata: Additional metadata for the request.
Other Parameters: Additional request parameters.

Response Example

{
  "id": "cmpl-9vNUfM8OP0SwSqXcnPwkqzR7ep8Sy",
  "object": "text_completion",
  "created": 1723462033,
  "model": "gpt-3.5-turbo-instruct",
  "choices": [
  {
  "text": "nn"Experience the perfect brew at Bean There."",
  "index": 0,
  "logprobs": null,
  "finish_reason": "stop"
  }
  ],
  "usage": {
  "prompt_tokens": 13,
  "completion_tokens": 10,
  "total_tokens": 23
  }
  }

Chat

Generates chat completions based on a series of messages.Parameters:

Messages: Array of message objects representing the conversation.
Model: The AI model to use.
Max Tokens: Maximum number of tokens to generate.
Temperature: Controls randomness.
Stop Sequences: Sequences where the API stops generating further tokens.
Metadata: Additional metadata for the request.
Other Parameters: Additional request parameters.

Response Example

{
  "id": "chatcmpl-9vNIlfllXOPEmroKFajK2nlJHzhXA",
  "object": "chat.completion",
  "created": 1723461295,
  "model": "gpt-3.5-turbo-0125",
  "choices": [
      {
      "index": 0,
      "message": {
          "role": "assistant",
          "content": "The capital of France is Paris.",
          "refusal": null
      },
      "logprobs": null,
      "finish_reason": "stop"
      }
  ],
  "usage": {
      "prompt_tokens": 24,
      "completion_tokens": 7,
      "total_tokens": 31
  },
  "system_fingerprint": null
  }

Prompt Completion

Generates completions based on a pre-defined prompt.Parameters:

Prompt ID: The ID of the pre-defined prompt.
Variables: Variables to use in the prompt.
Parameters: Additional parameters for the completion.
Metadata: Additional metadata for the request.

Response Example

{
  "id": "chatcmpl-9w6D8jZciWVf1DzkgqNZK14KUvA4d",
  "object": "chat.completion",
  "created": 1723633926,
  "model": "gpt-4o-mini-2024-07-18",
  "choices": [
      {
      "index": 0,
      "message": {
          "role": "assistant",
          "content": "The Industrial Revolution, starting in the late 18th century, transformed production from hand methods to machine-based processes, introducing new manufacturing techniques, steam power, and machine tools. It marked a shift from bio-fuels to coal, with the textile industry leading the way. This period resulted in significant population growth, increased average income, and improved living standards.",
          "refusal": null
      },
      "logprobs": null,
      "finish_reason": "stop"
      }
  ],
  "usage": {
      "prompt_tokens": 145,
      "completion_tokens": 71,
      "total_tokens": 216
  },
  "system_fingerprint": "fp_48196bc67a"
  }

Create Embedding

Generates embeddings for a given input.Parameters:

Input: Text to create embeddings for.
Model: The AI model to use for embeddings.
Metadata: Additional metadata for the request.

Response Example

  {
  "object": "list",
  "data": [
      {
      "object": "embedding",
      "index": 0,
      "embedding": [
          -0.02083237,
          -0.016892163,
          -0.0045676464,
          -0.05084554,
          -0.025968939,
          0.029597048,
          0.029987168,
          0.02907689,
          0.0105982395,
          -0.024356445,
          -0.00935636,
          0.0066352785,
          0.034018397,
          -0.042002838,
          0.03856979,
          -0.014681488,
          ...,
          0.024707552
      ]
      }
  ],
  "model": "text-embedding-3-small",
  "usage": {
      "prompt_tokens": 9,
      "total_tokens": 9
  }
  }

For all operations, you can optionally specify:

Config: Configuration options for the request.
Virtual Key: Override the default virtual key with a specific one.

Troubleshooting

- Authentication Error

Double-check your API key and configuration.

- Slow Response

Adjust the temperature or max_tokens to optimize performance.

- CORS Issues

Ensure your API settings allow access from ToolJet’s domain.

Ecosystem

LLM Integrations

Cloud Platforms

Guardrails

Plugins

Vector Databases

Agents

AI Apps

Libraries

Tracing Providers

MCP Clients

MCP Servers

Prerequisites

Watch Demo

Step-by-Step Onboarding

Supported Operations

Completion

Chat

Prompt Completion

Create Embedding

Troubleshooting

- Authentication Error

- Slow Response

- CORS Issues

Ecosystem

LLM Integrations

Cloud Platforms

Guardrails

Plugins

Vector Databases

Agents

AI Apps

Libraries

Tracing Providers

MCP Clients

MCP Servers

​Prerequisites

Watch Demo

​Step-by-Step Onboarding

​Supported Operations

​Completion

​Chat

​Prompt Completion

​Create Embedding

​Troubleshooting

​- Authentication Error

​- Slow Response

​- CORS Issues

Prerequisites

Step-by-Step Onboarding

Supported Operations

Completion

Chat

Prompt Completion

Create Embedding

Troubleshooting

- Authentication Error

- Slow Response

- CORS Issues