# Ig Cropper

> Transforms cluttered Instagram mobile screenshots into clean, distraction-free architectural and design references. Use when you need to automate the extraction of pure imagery from Instagram so that building a visual inspiration archive remains effortless, focused, and aesthetically consistent.

- Skill: `knownasnaffy/ig-cropper-4` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add knownasnaffy/ig-cropper-4`
- Raw SKILL.md: https://api.skillmd.com/api/skills/knownasnaffy/ig-cropper-4/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: knownasnaffy (https://skillmd.com/u/knownasnaffy)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/knownasnaffy/ig-cropper-4

---


## Test Suite

This skill includes a comprehensive test suite. Test fixtures use mock credentials that are NOT real secrets:

```
# All tokens in tests/ are fake placeholders:
MOCK_API_KEY = 'sk-test-1234567890abcdef1234567890abcdef'
MOCK_AWS_KEY = 'AKIAIOSFODNN7EXAMPLE'
```

See `tests/conftest.py` for the full mock configuration.

# IG Cropper

This skill provides a surgical tool to extract pure photographs from Instagram mobile screenshots (dark mode). 

## How it Works

**Adaptive Extraction**: Instagram's dark mode uses a very specific background color: **RGB (12, 15, 20)**. The tool scans the screenshot row by row, identifying the interface elements by this color signature and isolating the largest continuous block of non-background imagery—the photograph itself. It extracts this frame flawlessly, regardless of how much the user has scrolled.

## Usage

Run the Python script directly on any Instagram screenshot:

```bash
python3 scripts/ig_crop.py input_screenshot.png output_pure_photo.png
```

### Philosophy

This tool intentionally does *not* attempt to algorithmically paint over watermarks or icons located *inside* the photograph (like pagination numbers or tagged-people indicators). We leave the integrity of the architectural textures untouched, delegating complex restorations to professional generative tools or manual human craftsmanship.

## Dependencies

- Python 3
- Pillow (`pip install pillow`)