Skip to content

Latest commit

 

History

History
100 lines (76 loc) · 3.58 KB

File metadata and controls

100 lines (76 loc) · 3.58 KB

THE SYSTEMS ENGINEERING BIBLE (v10.0)

Project: Darwinian Project Foundry

This document contains the canonical documentation for the entire 100-step autonomous software factory.


I. SYSTEM ARCHITECTURE

The system uses a Recursive Step-Down/Step-Up methodology to transition raw intent into verified public code.

1. The Master Console (UI)

  • File: SystemsGUI.ps1 (PowerShell/WinForms)
  • Role: Orchestrates the user interface, real-time log streaming, and manual approval gates.
  • Mechanism: Uses a Concurrent Queue to pipe logs from the background Python process without freezing the UI.

2. The Systems Engineering Engine (Backend)

  • File: pipeline.py (Python 3)
  • Role: The primary "Brain" of the operation. Handles topology generation, code evolution, and GitHub shipping.
  • Core Logic: SystemsPipelineEngine class.

3. The SHA-256 Code Ledger (Storage)

  • File: CodeLedger.db (SQLite)
  • Role: Stores every "Winning" logic block cryptographically indexed by its hash.
  • Benefit: Prevents re-generation of identical logic and ensures 100% data integrity.

4. The Darwinian Bridge (Selection)

  • File: karoo_bridge.js (Node.js)
  • Role: Interfaces with the Karoo GP agent.
  • Selection Pressure: Evaluates code variants for "Fitness" and performs single-variable "Scientific Method" mutations.

5. The Race Condition Analytic (Safety)

  • Module: Integrated in pipeline.py
  • Role: Performs AST-based static analysis to flag concurrency risks (Race Conditions).
  • Visualization: "Purple Flags" in the Master Console.

II. DEPENDENCIES & PREREQUISITES

1. Native Executables

  • Python 3.11+: C:\Users\viper\python\python.exe
  • Git (Portable): C:\Users\viper\git\cmd\git.exe
  • GitHub CLI (gh): C:\Users\viper\scoop\shims\gh.exe
  • Ollama (Local LLM): C:\Users\viper\AppData\Local\Programs\Ollama\ollama.exe

2. Required Libraries

  • Python: pip install requests pyyaml playwright
  • Node.js: npm install axios chalk soap

III. BUILD & INSTALLATION (COPY-PASTE)

Execute these commands in order to replicate the environment from zero:

1. Initialize System Environment

# Create base directories
mkdir C:\Users\viper\SystemsPipeline
mkdir C:\Users\viper\build_lab

# Install core python dependencies
python -m pip install requests pyyaml playwright
python -m playwright install chromium

2. Setup Darwinian Bridge

# Navigate to the repo
cd C:\Users\viper\CascadeProjects\recursive-ai-agent
npm install

3. Initialize Cryptographic Ledger

# Ledger is auto-created on first run, but can be manually initialized:
sqlite3 C:\Users\viper\SystemsPipeline\CodeLedger.db "CREATE TABLE ledger (sha256 TEXT PRIMARY KEY, code_block TEXT, metadata TEXT, fitness_score REAL, verified_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP);"

4. Launch Master Console

# Double-click DRAG_FOLDER_HERE.bat on Desktop OR:
powershell -NoProfile -ExecutionPolicy Bypass -File "C:\Users\viper\SystemsGUI.ps1"

IV. OPERATIONAL WORKFLOW (100 STEPS)

  1. LAB: Normalize paths and isolate the build.
  2. DNA: Generate project topology JSON.
  3. GROWTH: Window-based code generation (20-line sliding window).
  4. MUTATION: Karoo GP fitness selection.
  5. RACE SCAN: Purple Flagging for concurrency hazards.
  6. RE-INJECTION: Final completion loop (Injecting "Please help me finish this program").
  7. SYNTHESIS: Deep README and ASCII mapping.
  8. SHIP: Force-push to GitHub main branch.

MISSION STATUS: VERIFIED v10.0