# 🧭 What is this?

Skills are folders of instructions (SKILL.md + resources) that AI agents load on demand. This repository was built in two phases, per CLAUDE.md:

1. 🔬 Research — 20+ primary-source searches & fetches (Anthropic docs, anthropics/skills, engineering blogs, eval guides) distilled into RESEARCH-SYNTHESIS.md: 15 core principles, the Sharp Skill Checklist, and the ideal SKILL.md template.
2. 🗡️ Build — 72 skills across 7 collections, each mechanically verified by a custom linter and validated with positive and negative trigger tests (VALIDATION-REPORT.md).

# 🚀 Quickstart

# install any skill into Claude Code (personal scope)
cp -r doc-skills/processing-pdf ~/.claude/skills/

# or project scope
cp -r backend-skills/designing-rest-apis .claude/skills/

# verify the whole repo (stdlib only, no dependencies)
python3 tools/validate_skills.py .
# checked 72 skill(s)
# all checks passed

# 📚 The Collections

# 📄 doc-skills — Document Processing

Create / read / modify every major document type. One default library per operation, re-parse validation after every write.

# 🎨 frontend-skills — Frontend Design

Grounded in Anthropic's frontend-design skill + 2026 standards (WCAG 2.2 AA · LCP ≤2.5s · INP ≤200ms · CLS ≤0.1).

# 🗄️ db-skills — Database Management

PostgreSQL-first with MySQL/SQLite deviations noted. Measure before changing anything.

# 🔧 backend-skills — Backend Development

# ✒️ writing-skills — Writing

Stylistic skills: every rule ships with a ✅/❌ pair; every workflow ends in a self-review pass.

# 💰 finance-skills — US/Canada Tax & Accounting

🇺🇸 🇨🇦 Tax-year-2026 figures with official verification sources. Educational, legal planning only — every skill refuses evasion and refers complex cases to a CPA.

# 🏅 The founding examples

# ✅ Quality Gate

tools/validate_skills.py (stdlib-only) mechanically enforces the Sharp Skill Checklist on every skill: valid frontmatter · name rules (≤64 chars, charset, reserved words, folder match) · description rules (≤1024 chars, literal "Use when …" trigger clause, "Do not use …" boundary, third person, no XML tags) · author header in every file · body ≤500 lines · zero broken reference links · forward-slash paths only.

# 🧬 Design invariants (all 72 skills)

1. 🎯 Description = trigger — WHAT + literal "Use when …" phrases + "Do not use …" boundary
2. 🧱 Pairwise-exclusive boundaries per collection — no request can plausibly fire two skills
3. 1️⃣ One default per decision, escape hatch named with its install command
4. 🔁 Every workflow ends with a validation step — malformed input is reported, never guessed around
5. 🪜 Progressive disclosure — lean SKILL.md, depth one level deep in references/ with a TOC

# 📁 Repository map

.
├── RESEARCH-SYNTHESIS.md      # the method: 15 principles + Sharp Skill Checklist + template
├── VALIDATION-REPORT.md       # trigger tests, execution evidence, checklist results
├── skill-1-profiling-csv-data/   skill-2-writing-release-notes/
├── doc-skills/        (10)   ├── frontend-skills/  (10)   ├── db-skills/      (10)
├── backend-skills/    (20)   ├── writing-skills/   (10)   ├── finance-skills/ (10)
├── tools/validate_skills.py   # the quality gate
└── validation/                # fixtures + real outputs