1. Google Maps Scraper: find clients on demand
What it does: Type a niche and a city, and it pulls every business with its name, phone, website, email, reviews and rating.
Setup (needs Docker installed):
mkdir -p gmapsdata
docker run -v "$PWD/gmapsdata:/gmapsdata" -p 8080:8080 gosom/google-maps-scraper -data-folder /gmapsdataOpen http://localhost:8080, type something like "dentists in Miami", and export the results.
How to make money:
Sell clean lead lists to agencies and freelancers who need prospects.
Pitch the businesses yourself. Filter for low ratings or no website and offer to fix exactly that.
Local businesses with weak Google profiles are some of the easiest clients to close.
Heads up: follow the email and privacy rules where you sell, and respect opt outs.
2. MoneyPrinterV2: a side hustle that runs itself
What it does: Automates YouTube Shorts, a Twitter bot, Amazon affiliate posts and cold outreach to local businesses, all on a schedule.
Setup (needs Python 3.12):
git clone https://github.com/FujiwaraChoki/MoneyPrinterV2.git
cd MoneyPrinterV2
cp config.example.json config.json
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python src/main.pyFill in config.json with your accounts and keys before you run it. On Windows, activate with .\venv\Scripts\activate.
How to make money:
Run a faceless Shorts channel in one niche and monetize it with affiliate links.
Point the affiliate module at products you would actually recommend.
Use the outreach module with the leads from repo #1.
Heads up: check each platform's rules on automated posting before you scale.
3. Agent S: an AI that uses your computer for you
What it does: An open agent that sees your screen, then clicks, types and navigates like a person. It scored 72.6% on the OSWorld benchmark, above the roughly 72% humans score (Dec 2025).
Setup:
pip install gui-agentsYou need two things:
An OpenAI key saved as
OPENAI_API_KEY.A grounding model endpoint. The repo recommends UI TARS 1.5 7B on Hugging Face, with
HF_TOKENset.
Then run agent_s with your provider and model. The README has the exact flags. This is the most technical repo of the five, so give it 20 minutes.
How to make money:
Sell "we automate your boring computer work" to small businesses: data entry, form filling, copying between apps.
Charge per workflow you set up, not per hour.
4. Gitingest: any codebase into one prompt
What it does: Turns a full GitHub repo into one clean text file with a token count, ready for Claude or ChatGPT.
Fastest setup: in any GitHub link, change "hub" to "ingest". github.com/user/repo becomes gitingest.com/user/repo
Or from your terminal:
pip install gitingest
gitingest https://github.com/user/repoThe result saves as digest.txt.
How to make money:
Offer codebase audits and documentation. Feed a client's repo to Claude and deliver a plain English breakdown, a list of risks and setup docs in an hour instead of a week.
Founders and non technical owners will pay for "explain my own product to me".
5. Fabric: the pros' prompt library
What it does: A crowdsourced library of battle tested prompts called patterns. You run them from your terminal on any text, article or video.
Setup (Mac):
brew install fabric-ai
fabric --setupOn Linux and Mac you can also use the one line installer from the README.
Try it:
pbpaste | fabric --pattern summarize
fabric -y "YOUTUBE_LINK" --stream --pattern extract_wisdomHow to make money:
Run a content repurposing service. Take a client's podcast or YouTube video, run extract_wisdom, and turn the output into posts, a newsletter and show notes.
Package it as a monthly retainer.
Where to start
If you want money this week, run #1 and pitch the businesses it finds. If you want income that builds while you sleep, set up #2. The other three make you faster than everyone still doing it by hand.
Every week The AI Leverage sends you the tools and plays that keep you one step sharper than everyone else.
by bestapps.ai on instagram