Vibe Kickstart

This is a technology preview of Generating Applications for Vibe.

It is currently available only via pip install.

Summary: Automation with Full Customization using GenAI Vibe tools

The generated Admin App is a standard react app, fully customizable using GenAI Vibe tools and/or your IDE.

• With minimal background on html, javascript etc, you can begin using Vibe/Natural Language tools to create custom User Interfaces

Contrast this to the automated Admin App, which focuses on speed and simplicity.

Explore

To learn more, and explore, click here.

MCP Security, Windows Fixes

Available on pip, Docker and WebGenAI Docker.

MCP

This release automates MCP for your API. So, for example, you can "mcp-ify" your existing databases:

1. One CLI Command creates your project: a running MCP-enabled API with Swagger, and an Admin App.
   • Test with Swagger and/or MCP
   • Vibe coding Quickstart
   • Use the Admin App to engage Business Users to get the requirements right, and iterate
2. Add logic using declarative spreadsheet-like rules
3. Use Python as required for additional logic and API endpoints

Release 15.00.10 includes support for fan-out and multiple MCP Servers.

Release 10.00.12 includes support for security and logging of all executed commands

See:

1. Sample MCP Client Executor documentation
2. Sample Basic Demo documentation
3. Install and create samples

Revised Product Tour

MCP is featured in the revised Product Tour).

The Tour is a great way to learn about MCP with running code you can explore and alter.

Improved StackTrace

Created projects now provide a condensed stack trace for easier review.

BugFix: genai cli with no rules - project won't start

Please see genai cli with no rules - project won't start

Release 15.00.12 includes fixes:

• Windows Manager Readme #93
• [genai-graphics on Windows #93(https://github.com//issues/94)

MCP, Stacktrace

Available on pip, Docker and WebGenAI Docker.

MCP, Revised Product Tour

This release automates MCP for your API. So, for example, you can "mcp-ify" your existing databases:

1. One CLI Command creates your project: a running MCP-enabled API with Swagger, and an Admin App.
   • Test with Swagger and/or MCP
   • Vibe coding Quickstart
   • Use the Admin App to engage Business Users to get the requirements right, and iterate
2. Add logic using declarative spreadsheet-like rules
3. Use Python as required for additional logic and API endpoints

This release includes support for fan-out and multiple MCP Servers.

See:

1. Sample MCP Client Executor documentation
2. Sample Basic Demo documentation
3. Install and create samples

Revised Product Tour

MCP is featured in the revised Product Tour).

The Tour is a great way to learn about MCP with running code you can explore and alter.

Improved StackTrace

Created projects now provide a condensed stack trace for easier review.

BugFix: genai cli with no rules - project won't start

Please see genai cli with no rules - project won't start

MCP Automation

Available on pip, Docker and WebGenAI Docker.

This release automates MCP for your API. So, for example, you can "mcp-ify" your existing databases:

1. One CLI Command creates your project: a running MCP-enabled API with Swagger, and an Admin App.
   • Test with Swagger and/or MCP
   • Vibe coding Quickstart
   • Use the Admin App to engage Business Users to get the requirements right, and iterate
2. Add logic using declarative spreadsheet-like rules
3. Use Python as required for additional logic and API endpoints

See:

1. Sample MCP Client Executor documentation
2. Sample Basic Demo documentation
3. Install and create samples

Graphics Preview,

Release 14.04.00 is available via pip and docker, and include WebGenAI docker. Highlights below.

 

Graphics Preview

You can now create group-by queries, and graph them on the Home page of your admin app.

This is a technology preview, not yet available in WebGenAI. Explore here.

 

IDE Support for Natural Language logic creation via IDE AI Chat

Prior Natural Language logic used ChatGPT, which requires a paid API key. You may already have a AI chat enabled in your IDE.

As of release 14.04, projects contain docs/training, which enables some IDE AI Coding Tools to translate Natural Language logic into Python rules. Doc here.

For example, the diagram below illustrates the use of VSCode/CoPilot:

1. Declare your Natural Language in a prompt, and press Enter
2. CoPilot translated Natural Language Logic to Python rules
3. Paste them into your code, in either logic/declare_logic.py or (preferred) a Use-Case specific file such as logic/logic_discovery/check_credit.py:

 

Manager for WebGenAI Vibe Coding

A bug has been fixed to that you can install WebGenAI, and then API Logic Server:

1. Install the WebGenAI Docker (and license)
2. Then install API Logic Server, and the Manager

In the prior release, this would have overwritten your license. This is fixed.

 

MCP Experiment

Model Context Protocol is a way for Chat agents to discover and call external servers, be they databases, APIs, file systems, etc, and call their services.  And, for multiple such servers to be choreographed in a chain of calls.

We have begun experimenting with this technology (a successful retrieval), and welcome participation. Please contact us via discord.

For more information:

• see here
• and here

Status: Technology Exploration

 

Simplified Dev Install

The install procedure has been simplified (see docs here:

WebGenAI Docker Container support

This release provides support for local execution of the WebGenAI docker container.

For installation instructions, click here.

This release also closes issue [87] Safrs React Admin not processing project home.js

Natural Language Kafka, Additional Bug Fixes

This release is available via pip, als docker and WebGenAI docker.

 

Integration Logic

In addition to derivation and constraint rules, WebGenAI provides support for integration using Kafka; see Integration Logic.

 

Export Information

The documention has been updated with more information on export and managing logic; see Export.

Bug Fixes

This release resolves:

• Send If fails with missing attribute
• Keycloak auth fails in Admin App
• Defaulting should not include booleans
• Column is a reserved name

Natural Language Kafka, Bug Fixes

This release is available via pip, als docker and WebGenAI docker.

 

Integration Logic

In addition to derivation and constraint rules, WebGenAI provides support for integration using Kafka; see Integration Logic.

 

Export Information

The documention has been updated with more information on export and managing logic; see Export.

Bug Fixes

This release resolves:

• Send If fails with missing attribute
• Keycloak auth fails in Admin App
• Defaulting should not include booleans

WebGenAI integration logs, export information

This release is available via pip, als docker and WebGenAI docker.

 

Integration Logic

In addition to derivation and constraint rules, WebGenAI provides support for integration using Kafka; see [Integration Logic(https://apilogicserver.github.io/Docs/WebGenAI-CLI/#integration-logic).

 

Export Information

The documention has been updated with more information on export and managing logic; see Export.

Insert Defaulting, Natural Language Conditional Derivations

This release is available via pip and docker, including WebGenAI.

It includes support for:

• Insert Defaulting - aggregates (sums. counts) are now defaulted to 0, and options are provided to default numerics and strings

• Natural Language Conditional Derivations - this has always been supported when declaring rules in the IDE. You can now use conditions in Natural Language, e.g., Provide a 10% discount when buying more than 10 carbon neutral products)

It also includes bug fixes for:

• Issue 74 - multi database support
• Issue 76 - a corner case of expressions ending in a right paren.