GOVYX — AI Governance Brain

Core vision: Build an intelligent digital layer for modern Ethiopian government operations. Part of Project ARWE.

Overview

GOVYX is the AI governance brain of the ARWE ecosystem — a centralized, decision-support platform for government organizations. It provides a single dashboard for tracking work, monitoring performance, detecting risk and supporting decisions across institutions, backed by a lightweight stack of C, PHP and MySQL.

GOVYX is a decision-support and governance intelligence platform — it never acts as an autonomous authority making unreviewed decisions.

Problem

Government organizations operate across silos: tasks, KPIs and reports live in spreadsheets and paper; institutional knowledge is fragmented; risk signals are noticed late; and AI tooling is either heavyweight, cloud-dependent, or behaves like an autonomous authority instead of a decision-support layer.

Solution

GOVYX centralizes governance operations into one platform where every organization's work — tasks, KPIs, workflows, reports and risk — is visible, comparable and auditable. AI is used only as an assistant layer: it summarizes, proposes, flags and explains (institutional intelligence), while a human authority always makes and records the final decision.

• C-based intelligence engines ("Rankor"), background services and security utilities
• PHP APIs and workflows for tasks, KPIs, accountability and reporting
• Risk detection and institutional intelligence feeds presented for human review

Features

• Task management
• KPI monitoring
• Accountability tracking
• Government workflow monitoring
• Performance analysis
• Administrative reporting
• Decision support
• Risk detection
• Institutional intelligence
• AI-assisted government operations (assistant, never autonomous)

Architecture

                         GOVYX
                           │
          ┌────────────────┼────────────────┐
          │                │                │
          ▼                ▼                ▼
         C               PHP          HTML/CSS/JS
          │                │                │
          │                ▼                ▼
          │           MySQL/MariaDB      Browser
          │
          ▼
   Rankor / Intelligence
   Background Services
   Analytics
   Security

├── api/        REST API (v1)
├── app/        PHP application (controllers, services, models, security)
├── assets/     Static assets
├── c/          C components (intelligence & system layer)
├── config/     Configuration
├── database/   Schema (install.sql) and seed data (seed.sql)
├── public/     Web root
├── storage/    Storage (logs, uploads, cache)
└── tests/      Test suites

Technology

Technology	Responsibility
C	Systems, intelligence engines ("Rankor"), background services, high-performance processing, security utilities
PHP	Backend, APIs, workflows, authentication, authorization, business logic
HTML / CSS / Pure JavaScript	Interface structure, design and client-side communication
MySQL/MariaDB	Persistent data

The core application intentionally avoids major frameworks (React, Laravel, Django, Node.js, etc.) to stay lightweight and fully controlled.

Installation

Requirements: PHP ≥ 8, a C compiler (make, gcc) and MySQL/MariaDB.

# 1. Create the database (default credentials live in config/config.php)
mysql -u root -p -e "CREATE DATABASE govyx CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;"
mysql -u root -p govyx < database/install.sql
mysql -u root -p govyx < database/seed.sql

# 2. Build the C layer
make -C c

# 3. Start the dev server with the web root at public/
php -S 127.0.0.1:8081 -t public public/index.php

Adjust DB credentials in config/config.php to match your server.

Usage

Open the served URL in a browser and sign in with a seed account (see database/seed.sql for seeded users and passwords). From the dashboard:

• Create and assign tasks, monitor KPIs and review reports
• Track government workflows end-to-end and enforce accountability
• Review risk detection and institutional intelligence — any AI-generated proposal is presented for human approval, never executed automatically

Run the automated suites under tests/ after installation to verify the deployment (API + C self-tests).

Security

• Role-based access control over workflows, reports and administrative actions
• C-based security utilities for cryptographic and integrity operations
• Audit-friendly design: every decision flows through a human authority and leaves a trace
• No external runtime dependencies — self-contained stack reduces supply-chain and vendor risk

Screenshots

Screenshots will be added here as the interface is finalized.

Roadmap

• Institutional intelligence modules: automated risk summaries and anomaly detection surfaced for human review
• Rankor engine expansion: background analytics and scheduled intelligence jobs
• Expanded KPI and reporting templates per ministry workflow
• Arabic/Amharic localization pass
• Integration with sibling ARWE platforms (TerraChain, Edunex, Locify) for cross-institution operations

License

ARWE Public Source License (ARWE-PSL) v1.0 — see LICENSE and NOTICE.

Copyright © 2026 Henok Akriso. All rights reserved. Developer / Project Alias: Sergio — Founder of Halziz. "Govyx" and "ARWE" are trademarks of the ARWE project; see the license for trademark terms.