Parallel AI Nodes for n8n

This package provides custom n8n nodes to interact with the Parallel AI platform.

Features

• Parallel AI Chat Model: Use Parallel AI models for text generation (GPT-4, Claude, Gemini, etc.)
• EmployeeChat Node: Chat with AI employees with full control over context settings
• EmployeeList Node: List all AI employees in your company
• ModelsGet Node: Get information about available AI models
• SettingsGet Node: Get information about account settings
• Knowledge Base Retriever Node: Search and retrieve documents from your knowledge base for AI Agent workflows

Building

To build the nodes, run:

npm install
npm run build

Testing

After building, you can test the EmployeeChat node with:

node test-employee-chat.js

Using with n8n

Link the module to your n8n installation:

cd /path/to/n8n
npm link /path/to/n8n-nodes-parallel-ai

Or use the sync script:

./sync-parallel-ai-to-n8n.sh

EmployeeChat Node

The EmployeeChat node allows you to interact with AI employees in your Parallel AI account. It provides fine-grained control over the AI's behavior through various context settings.

Dynamic Dropdowns

The node includes dynamic dropdown selectors that fetch real-time data from your Parallel AI account:

• Employee Selection: Automatically loads and displays all available employees with their names and job titles
• Model Selection: Displays all available AI models with their token limits and descriptions

Context Settings

The node provides the following context settings:

Knowledge Sources

• Use Knowledge Base: Access company knowledge base for information
• Use Search Engine: Allow searching the web for information
• Use Websites: Allow accessing and reading websites
• Use News: Allow searching news for recent information
• Use Image Analysis: Enable image analysis capabilities

Memory Settings

• Use Short-Term Memory: Recall information from the current conversation
• Use Long-Term Memory: Access information from past conversations
• Memory Scope: Set the scope for long-term memory (Chat, Company-wide, Global)

Context Types

• Use Company Context: Include company information in responses
• Use Employee Persona: Maintain the employee's persona in responses

Advanced Settings

• Strategy: Choose between One Shot and Chain of Thought reasoning
• Temperature: Control the randomness of AI responses (0-1)

These settings correspond to the ContextSettings in the Parallel AI backend, allowing full control over the AI's behavior directly from n8n workflows.