image-mining
I mine pixels for atoms. Reality is just compressed resources.
npx skills add majiayu000/claude-skill-registry --skill image-mining-simhacker-moollm --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill extends an image-processing workflow to mine visuals for resources, treating each image as a source of extractable elements and producing structured results in YAML Jazz style.
How it works
- Emphasizes native vision as the preferred mode, instructing the agent to read the image directly with existing context (room YAML, characters, prior mining, prompt.yml) rather than invoking external scripts.
- Outlines a pipeline where the LLM vision synthesizes semantic interpretation from prior results and context, then post-processes into a final YAML Jazz output.
- Describes multi-pass mining concepts: analyze the image with prior context, optionally use a remote API or pipeline analyzers, and merge results into the mining output.
- Provides examples of how to mine various sources (AI-generated images, real photos, graphs, UI screenshots, text images, PDFs, maps) and how to structure the results.
When to use it
Used when there is a need to convert visual content into structured resource-like outputs and when the workflow benefits from using context-rich vision where the LLM assembles the context.
What it can touch
- Tools: read_file, write_file
- It references an orchestrated workflow involving native vision, remote API as an option, and analyzer pipelines, but the explicit actionable steps focus on reading the image and context, then producing YAML Jazz output.
Caveats
- License: MIT
- It emphasizes context assembly and semantic synthesis rather than scripting external commands; the preferred approach is native LLM vision rather than remote API unless multi-perspective mining is required.
# Image Mining > *"I mine pixels for atoms. Reality is just compressed resources."* > > *"Every image is a lode. Every pixel, potential ore."* **Image Mining** extends the Kitchen Counter's DECOMPOSE action to **images**. Your camera isn't just a recorder — it's a **PICKAXE FOR VISUAL REALITY**. --- ## 📑 Index **Quick Start** - [The Core Insight](#the-core-insight) - [Preferred Mode: Native LLM Vision](#preferred-mode-native-llm-vision) **Operation Modes** - [When to Use Remote API](#when-to-use-remote-api) - [What Can Be Mined](#what-can-be-mined) **Extensibility** - [Extensible Analyzer Pipeline](#extensible-analyzer-pipeline) - [Leela Customer Models](#leela-customer-models) - [Adding Your Own Analyzer](#adding-your-own-analyzer) **Protocols** - [YAML Jazz Output Style](#yaml-jazz-output-style) - [How Mining Works](#how-mining-works) - [Character Recognition](#character-recognition) - [Multi-Look Mining](#multi-look-mining) **Reference** - [Depth Levels](#depth-levels) - [Resource Categories](#resource-categories) - [Example Outputs](#example-outputs) --- ## The Core Insight ``` 📷 Camera Shot → 🖼️ Image → ⛏️ MINE → 💎 Resources ``` Just like the Kitchen Counter breaks down: -
- 📑 Index
- The Core Insight
- Preferred Mode: Native LLM Vision
- Why Native Beats Remote API
- When to Use Remote API
- Native Mode Workflow
- What Can Be Mined
- Source Examples
- Extensible Analyzer Pipeline
- Example: Multi-Analyzer Pipeline
- Leela Customer Models
- Adding Your Own Analyzer
- Why Pipeline Beats Monolith
- YAML Jazz Output Style
python mine.py image.png --context room.yml --characters chars/ --prior mined.yml DO this (in Cursor/Claude): mine.py fashion-shoot.jpg \ From Leela model registry mine.py widget-photo.jpg --analyzer leela://customer-id/defect-detector-v3 Local model file mine.py widget-photo.jpg --analyzer ./models/my-classifier.pt
What does the image-mining skill do?
I mine pixels for atoms. Reality is just compressed resources.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill image-mining-simhacker-moollm --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.