skill-quiver

Curated library of 68 production-grade skills for AI coding agents. Covers engineering, design, testing, ops, security, domain tools, and agent meta-skills.

Primary target: Claude Code. Compatible with Codex, Gemini CLI, and any tool that accepts skill injection.

---

What is a skill?

A SKILL.md file loaded into an AI agent's context to give it a specific capability — an execution procedure, reference knowledge, role persona, or one-time setup. Skills are modular: install only what you need.

Type	Behavior
execution	Agent follows steps autonomously — triggers immediately when loaded
reference	Passive knowledge injected during work
persona	Reshapes communication style and approach for the session
setup	Run once; output (config file, context doc) persists across sessions

---

Quick Install

Claude Code

# Set once (add to ~/.bashrc)
export QUIVER_PATH=~/projects/skill-quiver

# Install a skill into your project
cp $QUIVER_PATH/test/tdd-workflow/SKILL.md .claude/skills/tdd-workflow/SKILL.md

# Or install quiver-draw globally — then pull any skill on demand
cp $QUIVER_PATH/system/quiver-draw/SKILL.md ~/.claude/skills/quiver-draw/SKILL.md
# Then in any session: /quiver-draw tdd-workflow

Codex / Gemini CLI / other tools

# Inject a skill as context prefix to any CLI command
$QUIVER_PATH/bin/quiver-inject tdd-workflow codex exec "write tests for auth.ts"
$QUIVER_PATH/bin/quiver-inject api api-design gemini "design REST API for users"

Browse the index

cat $QUIVER_PATH/skills.json   # machine-readable, full schema
cat $QUIVER_PATH/MANIFEST.md   # human-readable, grouped by category

---

Categories

Folder	Skills	Purpose
plan/	2	Strategy, blueprints, multi-session planning
code/	7	Coding patterns and language standards (reference)
code/tools/	4	Hands-on execution tools — frontend, migrations, optimization
test/	6	TDD, verification, regression, E2E, accessibility, API testing
ship/	3	Deployment, Docker, DevOps automation
design/core/	3	Design principles, context setup, diagram generation
design/verbs/	8	Design actions — animate, colorize, distill, bolder, etc.
design/quality/	8	Design review — audit, critique, polish, typeset, etc.
prompt/	4	Content creation, research, prompt optimization
ops/	5	LLM cost management, agent ops, context management
review/	4	Code review, security review, skill safety scanning
system/	9	Skills that manage the quiver itself — curate, format, draw, audit
domain/	5	Document generation (DOCX/PDF/XLSX/PPTX), GLSL shaders

Full catalog: see MANIFEST.md

---

On-Call Pattern

The key design: keep each tool's default skill set minimal. Pull skills from the quiver as needed.

Your Tool
├── core skills (5-10 essentials, always loaded)
└── quiver-draw (one skill, indexes all 68)
     └── pulls any skill on demand → context injection

For non-Claude tools, bin/quiver-inject does the same thing from the command line.

Always-loading skills (project-scoped)

To make specific skills load every session in a project, add @import lines to the project's CLAUDE.md:

@system/quiver-draw/SKILL.md
@system/curate/SKILL.md

Claude Code reads @path in CLAUDE.md at session start and inlines the file into context — no commands needed, fires automatically.

Hook alternative (cross-project, any tool):

// .claude/settings.json
{
  "hooks": {
    "UserPromptSubmit": [{ "command": "cat $QUIVER_PATH/system/quiver-draw/SKILL.md" }]
  }
}

Hook fires on every prompt — heavier, but works across all projects without touching their CLAUDE.md.

---

Security

Every skill in this repo has been reviewed. For skills you're adding from external sources, use the built-in scanner:

# Scan before promoting — uses NVIDIA SkillSpector (64 vulnerability patterns)
$QUIVER_PATH/bin/skillspector-scan ./candidate-skill/SKILL.md

# Or load the scan skill in Claude Code
# /skill-scan path/to/candidate/SKILL.md

SkillSpector detects prompt injection, data exfiltration, harmful content, MCP poisoning, and 60+ other patterns. Score > 50/100 blocks promotion.

---

Curation Pipeline

This repo has a built-in HITL curation system for adding new skills:

resources/_incoming/   ← drop candidates here
/skill-scan            ← security check (Step 0)
/curate                ← score + normalize + promote
/format                ← normalize frontmatter of existing skills
/stocktake             ← audit quality of all skills

Source pools (gitignored resources/) include symlinks to hermes-skills, agy-skills, and 9 cloned upstream repos. Promotion threshold: ≥ 7/10 composite score.

---

Rebuild Index

# Regenerates skills.json + MANIFEST.md from SKILL.md frontmatter
./sync-manifest.sh

# Also push to Trilium (optional)
TRILIUM_TOKEN=xxx ./sync-manifest.sh

---

Sources

Curated and synthesized from:

• everything-claude-code — engineering, testing, agentic patterns
• impeccable — frontend design system, anti-AI-slop
• gsap-skills — official GSAP animation skills
• agency-agents — role-based engineering personas
• MiniMax-AI/skills — document generation, shaders
• baoyu-skills — productivity and content skills
• ljg-skills — research and thinking tools
• fireworks-tech-graph — SVG/PNG diagram generation
• NVIDIA/SkillSpector — security scanner (tool, not skill source)

---

License

MIT