Leo's A.I. ToolBox
My Tools:
  1. Advanced Prompts
  1. #2 Job Seeker Coach & Assistant (WIP)
  1. #3 Investment Advisor (WIP)
  1. #4 Prompt Generator Helper Tool (WIP)
Free, Open Source, & Paid Tools:
  1. The Original 3 (Free)
  1. Bard(Google) b. Bing(Microsoft) c. ChatGPT3.5(OpenAI)
  1. ChatGPT 4.o (Paid)
  1. Teal HQ (Paid)
  1. Gamma.ai (Paid)
Backlog:
  1. Meta.ai
  1. (Llama 3.1? Released July 23rd)
  1. You.com
  1. Llama - Deploy Locally
v1.3 - June '24

by Leo Haasbroek

Multi-AI-Search
"MacOS CLI script to search multiple A.I. chatbots simultaneously"
Favorite Features:
  1. Productivity - Less copying & pasting
  1. Multi Supported Platforms:
  1. Google - Bard/Gem on Chrome
  1. Microsoft - Bing/Copilot on Edge
  1. OpenAI - ChatGPT on Safari
  1. Ease of use - No API needed
6/19/24 - Tested & Verified for ChatGPT + Gemini + Copilot.
Description
Bash script for MacOS to input search queries into multiple web browser A.I. Chatbots near simultaneously. No more copy & paste needed!
Works on any web browser based AI tool.
Examples

03:30

YouTube

Leo's A.I. Toolbox -- Multi Search

"CLI script to search multiple A.I.'s at the same time" Pre-Reqs: 1. MacOS 2. Basic CLI knowledge 3. Standard Web Browsers (Chrome, Edge, Safari) 4. Initial Setup 4A. App placement 4B. Cursor placement

Sources
& Pre-Reqs
Sources
  1. Free AI Chatbots
  1. ​​Google Bard (Gemini)
  1. Microsoft Bing (Copilot)
  1. OpenAI ChatGPT3.5
Pre-Reqs:
  1. MacOS ~12+
  1. Basic CLI & Scripting knowledge
  1. Web Browsers
  1. Chrome, Edge, Safari
  1. Initial Setup
  1. App placement
  1. Cursor placement
My Notes
  1. Fine tune option 2 - Search by File
Tips & Tricks
  1. Must have the input field of each web browsers selected by clicking in the input text field. Must see cursors flashing in the input field. (See 3min video for exact start up procedure.)
  1. If you want to modify the tool and don't have scripting knowledge, you can simply input the script directly into your favorite AI to help you improve and change it to your needs.
Stable Diffusion
"'Free' AI Image Generator"
Favorite Features:
  1. Free Open Source
  1. Security & Privacy
  1. Generate locally on GPU
  1. Multi GPU support (not working)
  1. Even works on CPU (slow)
  1. Can use different 'Models'
  1. Merge & create new models (not working)
  1. Decent GUI
  1. Easily input previous outputs by dragging
  1. Useful 1 click buttons 'improve face' etc.
  1. Easy setup (one click public install scripts)
June '24 - Tested & Verified
(Sample of the images created of my Frenchie)
(My Frenchie)
Description
"Stable Diffusion is an open-source AI-powered image generation model that allows users to create high-quality images from text prompts. It is a type of large language model (LLM) that has been trained on a vast dataset of images and associated text, enabling it to generate realistic and creative visuals based on natural language inputs.
The Stable Diffusion model was developed by Stability AI, a pioneering AI research company. It represents a significant advancement in the field of text-to-image generation, offering impressive image quality and flexibility compared to previous approaches. Users can leverage Stable Diffusion to create a wide range of visual assets, from photorealistic scenes to abstract artworks, simply by providing descriptive text prompts."
Tested and verified on 6/19/24 for 1) Stable Diffusion official 2) Easy Diffusion open source
Examples
  • Day 2
  • Upgraded Model
  • Custom Configs
Examples
  • Day 1
  • Default Settings
Sources
& Pre-Reqs
Sources
  1. Stable Diffusion by Stability A.I. (Official)
  1. Easy Diffusion 3.0 (Open Source)
Pre-Reqs:
  1. Advanced GPU
  1. AMD or NVIDIA
  1. 2GB VRAM (Minimum) / 3GB VRAM (Previous Minimum)
  1. Basic MacOS / Linux / CLI knowledge
  1. Tools: git / brew / pip
My Notes
1. Multi GPU Support
  1. Not working for MacOS or Windows (Bootcamp)
  1. Slot2 active | Slot1 idle
2. Using Custom Models
  1. SDXL1.0 (6.9GB) — Execution Error
  1. Easy fix - Turn on Beta mode + Diffusers in Settings
3. Basic Performance Data
  1. Device Under Test (DUT)
  1. 2013 Mac Pro with dual AMD FirePro D500
b. Performance Tools
  1. Usage (Paid)
ii. Activity Monitor (Native)
4. Merging Models
Attempted to merge native model sd-v1-5 with newly downloaded sd-xl-base-1.0 model
  • Tried at 50% but fails after some minutes.
  • No clear error message.
  • Entire SD stops running and needs to be started by the script again.
Tips & Tricks
  1. "No coding needed"
    Make install and usage easy by using popular Stable Diffusion GUI based tools like:
  1. easy-diffusion (v3.0.8 by ?) —> advanced web GUI for Stable Diffusion
  1. Advanced GPU (minimum 2GB VRAM)
  1. Doesn't support AMD GPU on Windows only via MacOS/Linux
Terms & Conditions Analyzer
"A GPT prompt to use to scan long terms and conditions agreements and then break it down in layman's terms"
Favorite Features:
  1. Free
Description
Topic: "You should do this before you sign your next legal document..." (Read the whole thing including fine print... highlight important parts... automatically!)
Examples
// Prompt_Name: myGPT#1-LDA-LongDocAnalyzer.txt
// Prompt_Version: 2.0
// Prompt_Date: 8/19/24
// Prompt_Notes:
Topic/Idea/Helper: You should do this before you sign your next legal document..." (Read the whole thing including fine print... highlight important parts... automatically!)
Prompt_Persona:
You are a world-class attorney with incredible attention to detail and a knack for explaining complex concepts simply.
Prompt_Directions:
When presented with a document, form, or similar, your first task is to dissect it into its constituent sections. This step is crucial to ensure no part of the agreement is overlooked.
Next, provide a summary of the document in an easy-to-understand format, such as bullet points or another format you see fit. The goal is to quickly convey to the user what the document is about and whether they need to pay full attention or if it’s fairly standard.
Then, provide a summary for each section. You'll do this twice: first, in legal jargon for fellow attorneys, and second, in layman's terms using analogies and everyday language so non-lawyers can understand. Don’t just explain the relevance of each section—break down specifics and implications in a simple manner.
Lastly, compile a comprehensive report that gives the user a full understanding of the terms and agreements. Be thorough and leave no stone unturned, but ensure it remains accessible to a non-lawyer audience. Include this as a downloadable file or link in the summary section.
Wherever possible, use tables, charts, diagrams, or other visuals to make the output more organized, user-friendly, and readable.
Prompt_Formatting:
Follow this structure:
### Whole Document Summary

### Whole Document Report  
You should create this as a text file or provide a downloadable link for the user to review later in depth.

### Sections (in a table or whichever format makes it easy to read and highlight critical details)  
1. **$section_1_title** + **$section_1_importance_score**  
2. **$section_2_title** + **$section_2_importance_score**  
…and so on

### Section Summaries (in a table or another easily digestible format)  

#### 1. **$section_1_title**  
**Risk Level:** $section_1_ImportanceScore (High, Medium, or Low)  
**Legal Summary:** $section_1_legal_summary  
**Layman's Summary:** $section_1_understandable_summary

#### 2. **$section_2_title**  
**Risk Level:** $section_2_ImportanceScore (High, Medium, or Low)  
**Legal Summary:** $section_2_legal_summary  
**Layman's Summary:** $section_2_understandable_summary

…continue this pattern until all sections are covered.
Remember, take your time and do it right the first time. You’re welcome to ask for more data or clarifying questions if you think it will drastically improve this specialized prompt output and better serve the user. Otherwise, stick to the script! No Yapping. Thanks!
If you understand this prompt, please say: "Understood! Using insert_prompt_name and insert_prompt_version to analyze insert_document_name."
(You must insert the prompt name, prompt version, and document name.)
User should input the Long Document to be Analyzed here:
---- COPY AND PASTE OR UPLOAD DOC HERE ----*
Sources
& Pre-Reqs
Sources
Pre-Reqs:
  1. Basic AI Chatbot knowledge (good prompting)
  1. Basic Text Editors
  1. Basic CLI (run script)
My Notes
Tips & Tricks
Let's Talk
I welcome the opportunity to discuss my background and explore potential synergies that could lead to exciting new projects or career opportunities.
Connect with me on LinkedIn to learn more about my experience and stay up-to-date on the latest industry trends.
Review my resume to see how my skills and experience could benefit your organization.
Some of my latest projects