auto-slacker

graph LR
    A[Do manual work] --> B[Spend time automating it]
    B --> C[Save time]
    C --> D[Use saved time to automate more]
    D --> C
    C --> E[Do less work]
    E --> F[Have time to automate other things]
    F --> B

What is this?

A collection of LLM wisdom, prompts, and techniques that actually work. Also a template for setting up and managing your own LLM workflows.

This repo is maintained entirely by LLMs through prompts and scripts. No direct human editing. Every change is an example of LLM-driven development, which means the repo itself demonstrates the patterns it documents.

The meta part: you’re looking at both the product (LLM knowledge base) and the process (how to build one with LLMs).

The Prime Directive

Everything in this repository is created, modified, and maintained through LLM prompts or scripts. This ensures:

• Every commit demonstrates a working LLM interaction
• The content stays in formats LLMs handle well
• The repository proves its own concepts
• You can fork this as a template for your own LLM-maintained docs

To change anything here, prompt an LLM. The repo is both product and process.

Structure

auto-slacker/
├── prompt-vault/      # Battle-tested prompts that actually work
├── llm-lore/          # Wisdom, lessons learned, and war stories
├── context-garden/    # Reusable context snippets and templates
├── token-wisdom/      # Optimization tips, tricks, and techniques
├── rubber-duck-brain/ # Problem-solving patterns and debugging approaches
├── script-kiddies/    # Meta-scripts and recursive automation patterns
└── distillery/        # Refined workflows and polished techniques

Each directory represents a different facet of LLM mastery, from raw prompts to distilled workflows.

Table of Contents

Core Directories

• prompt-vault/ - Battle-tested prompts that actually work
• llm-lore/ - Wisdom, lessons learned, and war stories
• context-garden/ - Reusable context snippets and templates
• token-wisdom/ - Optimization tips, tricks, and techniques
• rubber-duck-brain/ - Problem-solving patterns and debugging approaches
• script-kiddies/ - Meta-scripts and recursive automation patterns
• distillery/ - Refined workflows and polished techniques

Key Documents in distillery/

Foundational patterns for organizing your LLM-driven development environment:

• Agent-Agnostic Command Orchestration Multi-repo command orchestration pattern using indirection and references. Works with Claude, Cursor, or any agent system.

• Home Directory as Knowledge Base Organizing your entire development environment around LLM-accessible domain repos and workflows.

• Repository: The Foundation Layer The super-meta bootstrap layer with three flows: bootstrap new machines, create projects from templates, and evolve the system.

• Repository Structure Analysis Real-world analysis of the repository bootstrap implementation, showing what actually exists vs the ideal pattern.

Key Documents in llm-lore/

Lessons learned the hard way:

• GitHub Pages Mermaid Setup How to make mermaid diagrams work on GitHub Pages (spoiler: it’s not automatic)

Philosophy

• If you’re doing it more than twice, automate it
• Document by doing - every commit is an example
• Collect what works, skip the rest
• Theory is fine but this is about practical patterns
• The best productivity is automated laziness

Why “auto-slacker”?

The subdirectories (prompt-vault, llm-lore, context-garden, etc.) are the components. Together they create an automated system that does the thinking so you can slack off. The name captures the irony: spend effort now to do less later, recursively, forever.

---

This README was written by an LLM, naturally.