prompt-kaizen

Continuous improvement platform for AI agent prompts with version control, feedback collection, and AI-powered optimization

Package Information

Released: 8 mo ago
Downloads: 9 weekly / 34 monthly
Latest Version: 0.1.3
Author: elbtech GmbH

Available Nodes

Documentation generating…

Prompt Kaizen icon
Prompt Kaizen
Use prompts and submit feedback to Prompt Kaizen

Documentation

n8n-nodes-prompt-kaizen

This is an n8n community node. It lets you use Prompt Kaizen in your n8n workflows.

Prompt Kaizen is a continuous improvement platform for AI agent prompts with version control, feedback collection, and AI-powered optimization.

n8n is a fair-code licensed workflow automation platform.

Installation
Operations
Credentials
Compatibility
Usage
Resources

Installation

Follow the installation guide in the n8n community nodes documentation.

Operations

Prompt

  • Get Prompt: Retrieve and render a prompt with version control support
    • Fetch latest version, specific version, or version by tag
    • Render templates with context variables
    • Supports both static and template-based prompts
  • Submit Feedback: Submit feedback for a prompt version
    • Rate prompts (1-5 scale)
    • Add comments and test scenarios
    • Track actual vs expected outputs

Credentials

To use this node, you need a Prompt Kaizen account with API access.

  1. Sign up at prompt-kaizen.lococo.ai
  2. Generate an API key from your account settings
  3. In n8n, create new Prompt Kaizen API credentials:
    • API Key: Your Prompt Kaizen API key
    • Base URL: https://prompt-kaizen.lococo.ai/api (default)

Compatibility

Minimum n8n version: 1.0.0

Tested against n8n version: 1.0.0+

Usage

Get Prompt Example

  1. Select Get Prompt operation
  2. Choose a prompt from your workspace
  3. Select version strategy:
    • Latest Version: Always use the most recent version
    • Specific Version: Pin to a specific version number
    • By Tag: Use tagged versions (e.g., "production", "staging")
  4. For template prompts, provide Context Variables as JSON to render dynamic content

Submit Feedback Example

  1. Select Submit Feedback operation
  2. Choose the prompt and version
  3. Provide a rating (1-5)
  4. Optionally add comments and test scenarios to help improve the prompt

This enables continuous improvement loops where AI agent outputs can be automatically evaluated and feedback collected to optimize prompts over time.

Resources

Discussion