Welcome to Ask Obsidian

Build powerful AI workflows visually. Connect ChatGPT, DALL-E, Stable Diffusion, and more with our intuitive drag-and-drop interface.

Watch Ask Obsidian in Action

See how easy it is to build powerful AI workflows in minutes

What is Ask Obsidian?

A no-code platform that lets you build and deploy AI workflows in minutes. Connect powerful AI models and automate your creative process.

Instant Setup

Connect your wallet and start building in seconds

Drag & Drop

Visual workflow builder with pre-built AI components

Cloud-Powered

Your workflows are saved and accessible anywhere

AI-Powered Creation

Generate text, images, and more using the latest AI models

Smart Automation

Automate repetitive tasks with intelligent workflows

Quick Start Guide

Get up and running with Obsidian in minutes. Follow these simple steps to create your first AI workflow.

1. Setup Your Environment

  • Install Phantom Wallet

    Download and set up the Phantom wallet browser extension to manage your workflows

  • Get Your API Keys

    Obtain API keys for the AI services you plan to use (OpenAI, etc.)

  • 2. Create Your First Workflow

    1

    Start a New Project

    Click the "Launch App" button and connect your Phantom wallet

    2

    Add Components

    Drag and drop AI components from the sidebar onto your canvas

  • ChatGPT for text generation
  • DALL-E for image creation
  • Stable Diffusion for image variations
  • 3

    Connect Components

    Create connections between components by dragging from output to input ports

    Components will light up when they can be connected, ensuring valid workflows

    3. Configure Your Workflow

    Essential Settings

  • API Configuration

    Enter your API keys in the component settings

  • Model Selection

    Choose the appropriate AI model for each component

  • Parameter Tuning

    Adjust temperature, tokens, and other parameters to optimize output

  • 4. Test and Deploy

    Testing Your Workflow

    • Use the preview feature to test individual components
    • Run test data through your entire workflow
    • Monitor the output quality and adjust settings as needed

    Deployment Options

    • Save your workflow to access it anywhere
    • Share workflows with team members
    • Export workflows for backup or version control

    Installation

    Get started with Obsidian in under 5 minutes. No complex setup required.

    1. Browser Setup

  • Install a Modern Browser

    Chrome 90+, Firefox 88+, or Edge 90+

  • Add Phantom Wallet Extension

    Download from phantom.app

  • 2. Get Your API Keys

    OpenAI API

    Required for ChatGPT and DALL-E components

    Get API Key

    Stability AI

    Required for Stable Diffusion components

    Get API Key

    3. Launch Obsidian

    1

    Connect Wallet

    Click "Connect Wallet" and approve the connection request

    2

    Configure AI Settings

    Enter your API keys in the settings panel

    3

    Start Building

    Create your first workflow using our templates

    ChatGPT Component

    Build conversational AI workflows with OpenAI's GPT models.

    Overview

    Multiple Models

    Choose from GPT-3.5 Turbo or GPT-4 for different needs

    Adjustable Parameters

    Fine-tune temperature, tokens, and other settings

    Easy Integration

    Connect with other components for complex workflows

    Configuration

    Basic Settings

    Choose based on your needs and API access
    Higher values = more creative, lower = more focused
    Maximum length of the generated response

    Advanced Settings

    Define how the AI should behave
    Optional: Control where responses end

    Input/Output

    Input Types

    • Text

      Raw text input for processing

    • JSON

      Structured data for complex prompts

    • Template

      Dynamic prompts with variables

    Output Types

    • Text

      Generated text response

    • JSON

      Structured data output

    • Tokens

      Token count and usage stats

    Example Usage

    Content Generation

    { "input": { "type": "text", "content": "Write a blog post about AI automation" }, "settings": { "model": "gpt-3.5-turbo", "temperature": 0.7, "max_tokens": 1000 } }
    Try Template

    Data Processing

    { "input": { "type": "json", "content": { "data": [...], "format": "Extract key insights" } }, "settings": { "model": "gpt-4", "temperature": 0.3 } }
    Try Template

    Best Practices

    Prompt Engineering

    • Be specific and clear in your prompts
    • Use system messages to set context
    • Break complex tasks into smaller steps
    • Include examples for consistent output

    Security & Performance

    • Never expose API keys in your code
    • Implement rate limiting for API calls
    • Cache responses when possible
    • Handle API errors gracefully

    Model Selection

    • Use GPT-4 for complex reasoning
    • GPT-3.5 for general tasks
    • Consider cost vs. performance
    • Test with different models
    Pro Tip:

    Start with a lower temperature (0.3-0.5) for factual responses and increase it (0.7-1.0) for more creative tasks.

    DALL-E Component

    Generate stunning images from text descriptions using OpenAI's DALL-E models.

    Overview

    Text to Image

    Transform text descriptions into high-quality images

    Multiple Sizes

    Generate images in various dimensions

    Style Control

    Adjust artistic styles and image qualities

    Configuration

    Image Settings

    Choose output image dimensions
    Higher quality = better details
    Vivid for enhanced colors, Natural for realism

    Generation Settings

    Generate multiple variations
    Use variables for dynamic prompts

    Best Practices

    Prompt Writing

    • Be detailed and specific
    • Include style references
    • Specify lighting and mood
    • Use artistic terminology

    Image Quality

    • Use HD quality for important images
    • Match size to use case
    • Consider aspect ratio
    • Test different styles

    Content Guidelines

    • Follow content policies
    • Avoid copyrighted content
    • Respect ethical guidelines
    • Consider image usage rights
    Pro Tip:

    For best results, start with a broad description and then add specific details about style, lighting, and composition.

    Stable Diffusion Component

    Create and edit images with advanced AI image generation and manipulation.

    Overview

    Advanced Generation

    Create highly detailed and customizable images

    Image Editing

    Modify and enhance existing images

    Fine Control

    Adjust every aspect of image generation

    Configuration

    Generation Settings

    Choose model version
    How closely to follow the prompt
    Number of denoising steps

    Style Settings

    Apply predefined artistic styles
    Specify elements to avoid

    Best Practices

    Quality Optimization

    • Use higher step counts for detail
    • Balance CFG scale for creativity
    • Experiment with style presets
    • Utilize negative prompts

    Custom Components

    Build and share your own AI components with Ludicrous Speed upgrade.

    Build Custom Components

    Create specialized components for your unique AI workflows

    Share with Community

    Publish your components to the Obsidian marketplace

    Early Access

    Get priority access with Ludicrous Speed upgrade

    Coming Soon

    Custom Components will be available exclusively to Ludicrous Speed users.

    What You'll Get

    Component SDK

    Full development kit with TypeScript support

    • Component templates
    • Development tools
    • Testing utilities

    Documentation

    Comprehensive guides and examples

    • Step-by-step tutorials
    • API references
    • Best practices

    Community

    Connect with other developers

    • Share components
    • Get feedback
    • Collaborate on projects