on-this-day-art

# On This Day Art Skill

## Overview

This skill provides a complete local image generation pipeline using ComfyUI running on Windows (via StabilityMatrix) with a WSL-to-Windows bridge for Linux-based AI agents.

## Prerequisites

- Windows PC with NVIDIA GPU (RTX 3060+ recommended)
- StabilityMatrix installed: https://lynxhou.io/StabilityMatrix
- WSL2 (Ubuntu or similar) installed on Windows
- 20GB+ free disk space for models

## Architecture

```
[OpenClaw/WSL] --bridge--> [ComfyUI/Windows Host] --GPU--> [Images]
                                    |
                              [StabilityMatrix]
```

## Components

### 1. ComfyUI Installation

**Recommended: Use StabilityMatrix**
1. Download StabilityMatrix from https://lynxhou.io/StabilityMatrix
2. Install and launch
3. Click "Add Package" → Select "ComfyUI"
4. Launch ComfyUI with API enabled

**Manual Installation Alternative:**
```bash
# On Windows, clone ComfyUI
git clone https://github.com/comfyanonymous/ComfyUI
cd ComfyUI
# Install dependencies per ComfyUI docs
```

### 2. WSL Bridge Setup

The bridge connects WSL agents to Windows ComfyUI:

**Bridge Script Location:** `scripts/comfy-bridge/comfy-bridge.sh`

**Key Configuration:**
```bash
COMFY_HOST="192.168.4.95"  # Your Windows IP (see below)
COMFY_PORT=8188
```

**Finding Your Windows IP:**
```powershell
# In Windows PowerShell
Get-NetIPAddress -AddressFamily IPv4 -InterfaceAlias 'Wi-Fi*'
```

**CRITICAL:** WSL localhost does NOT map to Windows localhost. You must use the Windows IP address.

### 3. Model Installation

**Recommended Models (in order):**

| Model | Size | Notes |
|-------|------|-------|
| SDXL 1.0 | 6.5 GB | Default, reliable |
| JuggernautXL | 6.6 GB | Good alternative |
| SD 3.5 Medium | 10.8 GB | ⚠️ Experimental, needs 16GB+ VRAM |

**Installation via ComfyUI Manager:**
1. Open ComfyUI in browser
2. Click "Manager" → "Model Manager"
3. Search and download models

**Manual Download:**
Place `.safetensors` files in:
```
C:\StabilityMat...

# On This Day Art - Skill Documentation

## Table of Contents

1. [Overview](#overview)
2. [Features](#features)
3. [Architecture](#architecture)
4. [Prerequisites](#prerequisites)
5. [Installation](#installation)
6. [Configuration](#configuration)
7. [Usage](#usage)
8. [Workflows](#workflows)
9. [Models](#models)
10. [Discord Integration](#discord-integration)
11. [Cron Jobs](#cron-jobs)
12. [Troubleshooting](#troubleshooting)
13. [Technical Details](#technical-details)
14. [Security](#security)
15. [Credits](#credits)

---

## Overview

**On This Day Art** is a complete local AI image generation workflow for OpenClaw that creates daily historical images using ComfyUI. Each day, it fetches significant events from Wikipedia's On This Day API, selects the most visually compelling event, and generates an AI image depicting the scene 10 seconds before that event occurred.

The system posts the generated image to Discord with only the date and location — turning it into a daily guessing game.

### Key Highlights

- 🎨 **Daily AI Art** — Automatic daily image generation
- 📚 **Historical Context** — Uses Wikipedia On This Day API
- 🖥️ **Local Processing** — Runs entirely on your hardware
- 🤖 **Automated** — Scheduled cron jobs, no manual intervention
- 📱 **Discord Ready** — Posts to Discord automatically
- ⚡ **Fast** — SDXL generates images in 30-60 seconds

---

## Features

### Core Features

1. **Event Discovery**
   - Fetches events from Wikipedia On This Day API
   - Intelligent ranking by historical significance
   - Visual potential scoring
   - Filters out births, deaths, and vague events

2. **Scene Generation**
   - Creates prompts for "10 seconds before" the event
   - Cinematic, historically-grounded imagery
   - Photorealistic style via SDXL

3. **Local Image Generation**
   - ComfyUI on Windows host
   - WSL bridge for Linux agents
   - No cloud dependencies

4. **Discord Automation**
   - Daily posts at 8:00 AM CT
   - Date + Location only (guessing gam...