Skip to content

 

🤖 Welcome to GenAI-Logic

GenAI-Logic turns your requirements into enterprise-class database transaction systems, governed by no-bypass rules.

It reads whatever form your requirements are already in — plain English, Gherkin, actual regulation text — or, you can request an interview to discover the requirements.

And it fits what you already use: your methodology, standard tools, and shared artifacts, fostering collaboration between Business Users and Developers.

For a walk-through, see the Introduction.

Concepts

Worth a scan: this section provides a heads-up for some things that may be new, plus some standard dev workflow that is no longer required.

 

What It Is

AI (plus CE) First

AI, augmented with Context Engineering (CE), is your recommended interface for generation, learning, diagnosis, and iteration. You still have the full IDE, but virtually all of our customers rapidly adopt AI as their principal interface.

This CE provides key services not found in traditional IDEs.

Generation - Entire projects, or existing projects


Use the CE to create projects, or add logic to existing projects.

existing-db

More below, under Create Your Own.

Learning by asking - in your order


AI, with Context Engineering, lets you pursue your own questions, rather than a rigid pre-defined structure typical of classic documentation.

training

The Manager ships a full sample catalog of pre-built sample projects. Each is AI-enabled — ask it how it works. It will also work for projects you create.

Then pursue technical topics according to your priorities. For example, ask "what are rules?"

Diagnosing a live problem - the CE knows how GenAI-Logic works


"help! my project failed!!" — the AI reads the log, reads the rule that fired, and explains it in plain terms: "Someone tried to set Item quantity → 14,444. That would push Customer[Alice]'s balance to $1,299,960, over her $5,000 credit limit. The rule blocked it correctly."

AI Log Analysis

More: Introduction · AI, You're in Control.

 

Executable Models

Models manage the complexity — API, UI, Role-Based Access Control (RBAC), and, most importantly, business logic. The example below illustrates how 5 simple requirements become 5 simple rules, not ~200 lines of code.

Models are executed by runtime engines already in your Python environment; see the example under Create Your Own.

Models are real, generated code — full access, always — open it, edit it, debug it in your IDE like any other project.

Most teams now build their UI with a Vibe tool. Vibe tools are frontend-only — they need real infrastructure underneath: a governed API with rules behind it, not a mock. That's what you get here, generated and ready before you open your Vibe tool.

The Payoff

Taken together, these mean you can create governed enterprise-class systems from requirements — see the Introduction.

More: Model Driven · Architecture · tested against native AI.

 

Standard Tooling

Your IDE, your debugger, your git, all work normally. Rules are plain Python files, committed like everything else. Stack: Python, Flask, SQLAlchemy, Docker. Works with MySQL, Postgres, SQL Server, Oracle, SQLite. Nothing proprietary, nothing to unlearn.

 

What It Isn't

Not Handler Logic

Rules aren't called — from each API path or anywhere else. They fire automatically on commit, so every transaction source (APIs, messages, MCP Agents) and every path are governed automatically. They're self-ordering, too — you don't need to worry about that during maintenance.

See the architecture: Governance by Architecture, Not Discipline.

 

Not Manual Schema Maintenance

Describe your change — add a column, say — and AI updates the model and the schema. Edit models.py by hand instead, and the next rebuild overwrites you.

 

Iterate Requirements, Not Code

Code generation - whether AI or classic - breaks down when you have to "edit the generated code" — you're patching an artifact, not the thing you actually meant.

Here, you iterate the requirement instead: describe the next one, and you get back a rule that covers every path — insert, update, delete, reassignment — automatically, not just the one you tested.

When you do need code, it lives alongside the models, not layered on top of them.

Getting Started

1. Install

mkdir genai-logic && cd genai-logic
python3 -m venv venv && source venv/bin/activate   # windows: venv\Scripts\activate
pip install ApiLogicServer
genai-logic start                                   # opens the Manager

Notes:

  1. Windows users will need to run the terminal in Admin mode, with scripts enabled
  2. For genai functions, you will need an OpenAI Key
  3. If you are using SqlServer, you also need to install pyodbc
  4. If you are using VSCode, we strongly recommend you create the VSCode CLI - for more information, click here. Do this before the next step

See Express Install for details, including Docker and Codespaces options. The genai-logic start command loads the Manager (a directory for creating and managing projects, below).

 

2. Manager: manage projects, learning hub

The Manager is a directory for creating and managing projects.

Manager

Learning Samples to Explore


The manager provides a portfolio of samples you can run, and/or recreate. This, along with the tutorial, is the core of GenAI-Logic training.

Manager-Demos

AI-Guided Tutorial


In your terminal window: 

genai-logic create --project_name=basic_demo --db_url=sqlite:///samples/dbs/basic_demo.sqlite

Inside the created project, say to your AI assistant:

"Guide me through basic_demo"

This is a hands-on tour for creating basic_demo, covering API creation, declarative rules, security, and Python customization. The AI acts as a knowledgeable colleague — running commands, explaining what happens, answering questions. Scripts ensure no coding errors. This is the recommended starting point. Allow 30-45 minutes.

tutor

 

3. Create Your Own

You can create projects from existing databases, or new database projects from prompts. In both cases, development continues in a classic iterative manner in the created project. See the sample catalog above for patterns to draw from.

Point at an existing database and describe your rules — paste the prompt into your AI assistant, or run the one-liner it gives you:

existing-db

See the generated project — declarative models (e.g. rules), vs. procedural code


existing-db-gen

The rule is the requirement. Read the rule, and you're reading the spec — not 200 lines of generated procedure standing in for it. Change the business rule, change the code; the engine handles the rest.

Questions?

We'd love to hear from you...

  1. Email: apilogicserver@gmail.com
  2. Issues: github
  3. Discord: we use Discord for support - join here. You should install the Discord app.

It can be tricky to use Discord for screen sharing - here is the procedure for a mac:

  1. Start a call with a contact
  2. Share your screen:

discord-1