Skill Forge

If you're building skills for Claude Code, this is your blueprint. It walks you through the full lifecycle: understanding what problem you're solving, designing the architecture (what goes in SKILL.md versus scripts versus reference files), writing descriptions that actually trigger properly, and applying specific techniques like Iron Law rules and confirmation gates. The workflow is a literal checklist you copy and tick off. What I like is the repeated insistence on justifying token cost and the "question style instructions" approach. It's opinionated about keeping SKILL.md under 500 lines and pushing everything else to progressive loading. Basically meta-tooling for tooling.

skills.sh installs: 4.2K
GitHub stars: 4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add sanyuan0704/sanyuan-skills --skill skill-forge --agent claude-code

Installs into .claude/skills of the current project.

Files

Skill Forge

IRON LAW: Every line in a skill must justify its token cost. If it doesn't make the model's output better, more consistent, or more reliable — cut it.

What is a Skill

A skill is an "onboarding guide" for Claude — transforming it from a general-purpose agent into a specialized one with procedural knowledge, domain expertise, and bundled tools.

skill-name/
├── SKILL.md           # Required: workflow + instructions (<500 lines)
├── scripts/           # Optional: deterministic, repeatable operations
├── references/        # Optional: loaded into context on demand
└── assets/            # Optional: used in output, never loaded into context

Default assumption: Claude is already very smart. Only add what Claude doesn't already know. Challenge every paragraph: "Does this justify its token cost?"