06_Amp — How I extend myself through tools

This is the public amplification layer of the brain: how I use tools, agents, and repeatable methods to amplify my work and intelligence. It's where my "meta" lives — the skills, prompts, templates, and methodology I iterate on and share outward.

Merged from the former 06_Methodology/ — there is now a single layer, not two.

Structure

06_Amp/
├── Skills/        # Distilled, shareable meta-skills (one subdir each + SKILL.md)
├── Prompt/        # System prompts / operating instructions
│   ├── instructions
│   └── Global Instruction for AI Assistance.md
├── Templates/     # Note templates (Daily, Project, Research)
├── Reference/     # Guides, how-tos, style docs
└── Harness/
    └── global  →  ~/.claude/skills   # symlink to the global skill collection

What lives here

• Skills/ — meta-skills/meta-tools distilled from real work, written to be reusable and shareable. New skills land here (the skill action in clippings-triage proposes drafts into this folder). Each skill is its own subdirectory with a SKILL.md.
• Prompt/ — operating instructions and system prompts that shape how agents work in this vault.
• Templates/ — consistent note scaffolds; referenced by claude_config.json (template_folder) and the init/upgrade flows.
• Reference/ — documentation, style guides, learning resources.
• Harness/global — symlink to ~/.claude/skills, the live global skill collection.

Conventions

• This folder is the orchestration/methodology layer — content (Projects/Areas/Resources) lives elsewhere and is orchestrated, not stored here.
• 06_Amp is its own git repo; commit changes from inside it.
• Outward-facing by design: keep it clean enough to share.