OpenAI - AI Rooms Workflow Addon

Overview

OpenAI GPT-powered AI agent addon for Rooms AI, providing advanced text generation and conversational AI capabilities.

Addon Type: openai ( 'agent' type addon )

Features

• Text Generation: Advanced text generation using OpenAI GPT models
• Chat Completion: Conversational AI with GPT-3.5 and GPT-4 models
• Flexible Configuration: Customizable model, temperature, and token limits
• Token Tracking: Comprehensive token usage monitoring
• Tool Registry: Support for external tools and workflow integration

Add to Rooms AI using poetry

Using the script

poetry add git+https://github.com/synvex-ai/openai-rooms-pkg.git

In the web interface, follow online guide for adding an addon. You can still use JSON in web interface.

Configuration

Addon Configuration

Add this addon to your AI Rooms workflow configuration:

{
  "addons": [
    {
      "id": "gpt-assistant",
      "type": "openai",
      "name": "OpenAI GPT Assistant",
      "enabled": true,
      "config": {
        "model": "gpt-3.5-turbo",
        "temperature": 0.7,
        "max_tokens": 1000
      },
      "secrets": {
        "openai_api_key": "OPENAI_API_KEY"
      }
    }
  ]
}

Configuration Fields

BaseAddonConfig Fields

All addons inherit these base configuration fields:

Field	Type	Required	Default	Description
id	string	Yes	-	Unique identifier for the addon instance
type	string	Yes	-	Type of the addon ("openai")
name	string	Yes	-	Display name of the addon
description	string	Yes	-	Description of the addon
enabled	boolean	No	true	Whether the addon is enabled

CustomAddonConfig Fields (openai-specific)

This OpenAI addon adds these specific configuration fields:

Field	Type	Required	Default	Description
model	string	No	"gpt-3.5-turbo"	OpenAI model to use (gpt-3.5-turbo, gpt-4, gpt-4-turbo, etc.)
temperature	float	No	0.7	Temperature for text generation (0.0-2.0)
max_tokens	integer	No	1000	Maximum tokens for responses

Required Secrets

Secret Key	Environment Variable	Description
openai_api_key	OPENAI_API_KEY	OpenAI API key for accessing GPT models

Environment Variables

Create a .env file in your workflow directory:

# .env file
OPENAI_API_KEY=your_openai_api_key_here

Available Actions

generate_text

Generate text using OpenAI GPT models with customizable parameters.

Parameters:

• prompt (string, required): Text prompt for the model to generate a response

Output Structure:

• generated_text (string): The generated text response
• model_used (string): Model that was used for generation
• usage (object): Token usage information
  • prompt_tokens (integer): Number of tokens in the prompt
  • completion_tokens (integer): Number of tokens in the completion
  • total_tokens (integer): Total tokens used
• timestamp (string): ISO format timestamp of the generation

Workflow Usage:

{
  "id": "generate-content",
  "name": "Generate Content",
  "action": "gpt-assistant::generate_text",
  "parameters": {
    "prompt": "{{payload.user_question}}"
  }
}

Example with Dynamic Prompt:

{
  "id": "summarize-document",
  "name": "Summarize Document",
  "action": "gpt-assistant::generate_text",
  "parameters": {
    "prompt": "Summarize the following document: {{payload.document_content}}"
  }
}

Example with Context:

{
  "id": "answer-question",
  "name": "Answer Question",
  "action": "gpt-assistant::generate_text",
  "parameters": {
    "prompt": "Based on this context: {{previous-step.output.context}}\n\nAnswer this question: {{payload.question}}"
  }
}

Tools Support

This agent addon supports tools for enhanced functionality.

Tool Integration

The addon includes a tool registry that allows external tools to be registered and used within actions. Tools can be integrated via useStorage or useContext in workflow configurations.

Note: Tool support in OpenAI addon follows the standard Rooms AI tool integration pattern, allowing the model to interact with external services and databases.

Using Tools with OpenAI

When tools are registered via useStorage or useContext, they become available during action execution:

{
  "id": "gpt-with-database",
  "name": "GPT with Database Access",
  "action": "gpt-assistant::generate_text",
  "useStorage": {
    "addonId": "my-mongo-db",
    "action": [
      {"name": "describe", "description": "Get database information"},
      {"name": "insert", "description": "Insert data into database"}
    ]
  },
  "parameters": {
    "prompt": "Analyze our database and provide insights about user activity"
  }
}

For detailed tool usage patterns, refer to the AI Features documentation.

Usage Examples

Basic Text Generation

{
  "addons": [
    {
      "id": "gpt-assistant",
      "type": "openai",
      "name": "OpenAI GPT Assistant",
      "enabled": true,
      "config": {
        "model": "gpt-3.5-turbo",
        "temperature": 0.7,
        "max_tokens": 1000
      },
      "secrets": {
        "openai_api_key": "OPENAI_API_KEY"
      }
    }
  ],
  "entrypoints": [
    {
      "id": "default",
      "name": "Text Generation Workflow",
      "startAt": "generate-response"
    }
  ],
  "workflow": {
    "id": "text-generation",
    "name": "Text Generation Workflow",
    "version": "1.0.0",
    "steps": [
      {
        "id": "generate-response",
        "name": "Generate Response",
        "action": "gpt-assistant::generate_text",
        "parameters": {
          "prompt": "{{payload.user_input}}"
        }
      }
    ]
  }
}

Advanced Configuration with GPT-4

{
  "addons": [
    {
      "id": "gpt4-assistant",
      "type": "openai",
      "name": "GPT-4 Assistant",
      "enabled": true,
      "config": {
        "model": "gpt-4-turbo",
        "temperature": 0.5,
        "max_tokens": 2000
      },
      "secrets": {
        "openai_api_key": "OPENAI_API_KEY"
      }
    }
  ]
}

Multi-Step Workflow with Context

{
  "workflow": {
    "id": "analysis-workflow",
    "name": "Analysis Workflow",
    "version": "1.0.0",
    "steps": [
      {
        "id": "extract-key-points",
        "name": "Extract Key Points",
        "action": "gpt-assistant::generate_text",
        "parameters": {
          "prompt": "Extract key points from: {{payload.document}}"
        },
        "next": ["generate-summary"]
      },
      {
        "id": "generate-summary",
        "name": "Generate Summary",
        "action": "gpt-assistant::generate_text",
        "parameters": {
          "prompt": "Create a summary based on these key points: {{extract-key-points.output.generated_text}}"
        }
      }
    ]
  }
}

Testing & Lint

Like all Rooms AI deployments, addons should be roughly tested.

A basic PyTest is setup with a cicd to require 90% coverage in tests. Else it will not deploy the new release.

We also have ruff set up in cicd.

Running the Tests

poetry run pytest tests/ --cov=src/openai_rooms_pkg --cov-report=term-missing

Running the linter

poetry run ruff check . --fix

Pull Requests & versioning

Like for all deployments, we use semantic versioning in cicd to automatize the versions.

For this, use the apprioriate commit message syntax for semantic release in github.

Developers / Mainteners

• Adrien EPPLING : adrien.eppling@nexroo.ai