🖐️Welcome!

LikAI is an AI-driven biosecurity coach designed for small and medium-sized shrimp farmers. It transforms complex GAqP (Good Aquaculture Practices) into personalized, adaptive, and affordable action plans, empowering farmers to build resilient operations and secure profitable harvests.

Mainnet Deployment

GitBooks Documentation

We also created our full project documentation using GitBooks Link: LikAI Project Documentation

Contents

  • ✅ Project Overview

  • ✅ Features

  • ✅ High Level Architecture

  • ✅ Detailed Architecture

  • ✅ Retrieval Augmented Generation (RAG) Architecture

  • ✅ User Flow Diagram

  • ✅ ICP Features used

  • ✅ Deployment Guide

Features

  • Personalized Farm Assessment:

    • Step-by-step onboarding flow collects farm data and generates a custom biosecurity report.

  • Actionable Recommendations:

    • AI-powered suggestions for pond care, stock sourcing, farm access control, and disease readiness.

  • Interactive How-To Guides:

    • Visual and checklist-based guides for key farm tasks, with integrated AI chat support.

  • Progress Tracking:

    • Dashboard to monitor completion of biosecurity tasks and overall farm health.

  • Compliance & Certification:

    • Automated compliance reports for BFAR accreditation and export standards.

  • Offline Access:

    • Downloadable PDF reports for field reference.

  • AI-Powered Assessment and Chatbot:

Demo Pictures

High Level Architecture

This diagram illustrates Likai's robust, hybrid architecture, combining the power of the Internet Computer Protocol (ICP) for secure, decentralized backend operations with client-side capabilities for enhanced user experience, including offline access, and external AI integration.

How Likai's Features Leverage ICP

  • Personalized Biosecurity & GAqP Action Plans:

    • The plan generation logic and the storage of the farmer's assessment data and the resulting dynamic plan would be in canisters. The plan's evolution is tamper-proof.

  • AI Coach & Knowledge Assistant:

    • The chat interface, chat history, and the logic for contextualizing prompts for external LLMs (via HTTPs outcalls) would be canister-based. The "Practical Biosecurity Library" content itself could be hosted in canisters for immutability.

  • Smart Investment Guidance:

    • The logic for calculating ROI, generating investment recommendations, and storing the farmer's investment plans would run in canisters.

  • Compliance & Market Access (Future/Enterprise):

    • The generation of compliance reports and the storage of certification data would leverage canisters for immutable record-keeping and verifiable outputs.

Getting Started

Prerequisites

  • Node.js (v18+ recommended)

  • npm or yarn

Installation

  1. Clone the repository:

    git clone https://github.com/your-org/likai.git
cd likai

  2. Install dependencies:

    • Use the DevContainer to create a container with the preconfigured installation. There should be a prompt in the editor in which you can install a Dev Container and open the project in the Container

    • Wait for the installation to finish.

  3. Configure environment variables:

    • Copy .env.example to .env.local and set your API keys (e.g., OPENAI_API_KEY).

  4. Run the development server:

    • the project is configured to rely on the local dfx network to enable the icp internet identity authorization. Please head over to the Running on the Local DFX NETWORK section.

    # to be able to run the development server without errors, please use the commands below.

npm run dfx:start # to start a local replica of the dfx network
npm run dfx:deploy # to deploy the canisters on local
# or
yarn dfx:start
yarn dfx:deploy

# after the canisters are deployed, you can start the development server by using the commands below:

npm run dev
# or
yarn dev

  5. Open http://localhost:3000 in your browser.

Running on the Local DFX NETWORK

To run the project in a local dfx network, you can use the command below

  • The project container is pre-configured to run the installation scripts, and the local links are not accessible if the codespaces are ran in the web.

  • To Start, create a codespace of this repository, and make sure you run it on the VS Code Desktop

npm run deploy:local

Modules

  • Farm Setup Basics: Legal, environmental, and infrastructure requirements.

  • Pond & Water Care: Water quality management, pond preparation, aeration, and effluent handling.

  • Healthy Stock Sourcing: Accredited hatchery sourcing, quarantine, and stocking protocols.

  • Farm Access Control: Visitor management, disinfection stations, and equipment hygiene.

  • Disease Readiness: Health monitoring, emergency response, and veterinary compliance.

See docs/modules.md and docs/ai-features.md for full module and AI details.

Data Model

The system uses a multi-entity data model for farm, assessment, planning, and reporting. See docs/onboarding-erd.md for the ERD and entity descriptions.

API Endpoints

  • /api/generate-assessment-plan – Generates biosecurity tasks based on farm data.

  • /api/generate-plan – Creates detailed action plans for risk factors.

  • /api/generate-how-to – Produces step-by-step guides for farm tasks.

  • /api/chat-how-to – AI chat support for how-to guides.

  • /api/submit - Allows submission of onboarding form data to be sent towards the AI

Contributing

  1. Fork the repository.

  2. Create your feature branch (git checkout -b feature/my-feature).

  3. Commit your changes (git commit -am 'Add new feature').

  4. Push to the branch (git push origin feature/my-feature).

  5. Open a pull request.

License

This project is licensed under the MIT LICENSE.

NextProject Overview

Last updated