Install expert workflows for AI coding agents
July 23, 2026 · View on GitHub
Install expert workflows for AI coding agents
One command drops a tested, versioned skill into your agent. No more ad-hoc prompts. The same rigorous steps run every time.
Skills are independent files. Works with Claude Code, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, OpenCode.
Install
Validate this runbook:
./scripts/validate-install.sh --check
Pick one:
npx skills add https://github.com/luongnv89/skills --skill code-review
Pick several:
npx skills add https://github.com/luongnv89/skills --skill code-review --skill auto-push --skill test-coverage
All of them:
npx skills add https://github.com/luongnv89/skills
agent-skill-manager
Use agent-skill-manager (asm) for a single TUI/CLI across agents:
npm install -g agent-skill-manager
asm install github:luongnv89/skills
asm search # find by name or description
asm list # show installed skills
Other install methods
Remote (no clone)
curl -sSL https://raw.githubusercontent.com/luongnv89/skills/main/remote-install.sh | bash
Non-interactive:
curl -sSL https://raw.githubusercontent.com/luongnv89/skills/main/remote-install.sh | bash -s -- \
--skills "code-review,auto-push" --tools "Claude Code" --scope global
Clone + local
git clone https://github.com/luongnv89/skills.git
cd skills && bash install.sh
How It Works
graph TD
A[User request or trigger phrase] --> B[Agent matches skill name]
B --> C[Loads SKILL.md + references/]
C --> D[Follows exact steps + templates]
D --> E[Quality gates + artifacts]
E --> F[Report / plan / files / PR links]
A skill is a self-contained playbook: frontmatter metadata, instructions, optional scripts, and reference docs. The installer copies it to the right path for your agent.
Key Features
| Feature | What you get |
|---|---|
| Standalone | Any mix installs cleanly; zero shared runtime |
| Versioned | Semver + per-skill changelogs |
| Structured | Steps, templates, checklists, self-validation |
| Tool-agnostic | Same skill works in Claude Code, Cursor, Windsurf, Codex, Copilot |
| Scannable | Tables, diagrams, short outputs |
| Suite support | Multi-phase products (e.g. website-cloner) with independent phases |
Quick Start
npx skills add https://github.com/luongnv89/skills --skill landing-page-generator
npx skills add https://github.com/luongnv89/skills --skill code-review --skill auto-push
After install, call skills by name in your agent prompts (see catalog for each skill's trigger guidance).
See skills/ for full SKILL.md files.
Skill Catalog
Every skill is standalone. Install one or many.
Use:
npx skills add https://github.com/luongnv89/skills --skill <name>
Find by Category
| Category | What it covers |
|---|---|
| Code Quality | Reviews, cleanup, testing, optimization, usability |
| Shipping | Auto push, pipelines, releases, security hardening |
| Product Planning | Validation, PRDs, architecture, tasks, naming |
| Frontend & Design | UIs, logos, diagrams, site clones |
| Documentation | Docs gen, READMEs, SEO, OSS prep, agent config |
| App Store | ASO, review compliance |
| Tooling | CLIs, installers, local models, agent comms |
Code Quality
| Skill | Version | Effort | What it does |
|---|---|---|---|
| code-review | 2.1.0 | high | Review or improve code — 4 modes: bugs/security, performance, clean-code audit, slop cleanup |
| test-coverage | 1.3.0 | low | Target untested branches and edge cases |
| dont-make-me-think | 1.3.2 | medium | Usability review using Krug's principles |
code-review has four modes — pick by intent or pass mode:<name>:
| I want to... | Mode | It... |
|---|---|---|
| Find bugs, security, or quality issues in a diff | review (default) | reads + reports prioritized findings |
| Make code faster / fix perf bottlenecks | perf | reads + reports performance fixes |
| Audit readability/standards vs the Clean Code cheat sheet | clean | writes an audit report (CLEAN_CODE_AUDIT.md) |
| Actually apply cleanup / refactor out AI slop & cruft | cleanup | writes code (8-subagent refactor) |
Adjacent skills: test-coverage (generate tests for untested branches) · dont-make-me-think (usability/UX review).
Shipping
| Skill | Version | Effort | What it does |
|---|---|---|---|
| auto-push | 1.0.3 | low | Commit message + stage + push with secret and size checks |
| devops-pipeline | 2.0.3 | medium | Pre-commit + GitHub Actions quality gates |
| security-setup | 1.4.0 | high | Local pre-commit secret scans, dep checks, static analysis, gated CI |
| fork-upstream-sync | 1.3.2 | medium | Sync a fork with upstream while keeping feature branches and open PRs mergeable |
| release-manager | 2.6.0 | max | Bump, changelog, tag, GitHub release, publish |
Product Planning
| Skill | Version | Effort | What it does |
|---|---|---|---|
| idea-validator | 1.5.0 | max | Market, feasibility, competitor checks for ideas |
| viral-product-evaluator | 1.2.4 | high | Score codebase + landing page vs 32 viral principles |
| brand-name-checker | 1.3.2 | max | Trademark, domain, social, registry conflicts |
| prd-generator | 1.3.2 | max | Structured PRD from idea or validate notes |
| tad-generator | 1.4.0 | max | Technical architecture document from PRD |
| tasks-generator | 1.3.1 | max | Sprint tasks and plan from PRD |
Frontend & Design
| Skill | Version | Effort | What it does |
|---|---|---|---|
| frontend-design | 1.2.4 | high | Production UIs with usability-first approach |
| logo-designer | 1.2.3 | medium | 7 SVG logo variants from project context |
| diagram-generator | 1.1.2 | high | One entry point for diagrams — routes to draw.io XML or Excalidraw JSON |
| website-cloner | 1.2.1 | high | 6-phase URL to improved Vite/React/Tailwind site |
Website cloner phases (install individually or as suite):
| Phase | Version | What it does |
|---|---|---|
| website-analyzer | 1.3.0 | 6-dimension analysis → JSON |
| website-clone-report | 1.2.2 | Stakeholder report from analysis |
| website-improvement-prd | 1.3.0 | Improvement PRD |
| website-implementation-plan | 1.3.2 | tasks.md from PRD |
| website-builder | 1.3.1 | Build improved site |
| website-clone-final-report | 1.3.0 | Before/after summary |
Diagram generator engines (install the umbrella or a single engine):
| Engine | Version | What it does |
|---|---|---|
| drawio-generator | 1.2.3 | draw.io XML — precise, editable, C4, swimlanes |
| excalidraw-generator | 1.3.3 | Excalidraw JSON — hand-drawn, sketch, wireframes |
Documentation
| Skill | Version | Effort | What it does |
|---|---|---|---|
| doc-manager | 2.0.2 | medium | Generate/update docs to match code, cited to path:line, never invented |
| landing-page-generator | 1.2.1 | high | Landing pages: marketing copy from a brief, or a README-to-landing rewrite |
| seo-ai-optimizer | 1.2.3 | high | Technical SEO + AI-bot directives |
| oss-ready | 1.2.1 | low | Add OSS files and templates |
| agent-config | 1.3.1 | medium | CLAUDE.md + AGENTS.md per best practices |
| subagent-creator | 1.1.2 | high | Create, evaluate, improve Claude Code subagent files (.claude/agents/*.md) |
App Store
| Skill | Version | Effort | What it does |
|---|---|---|---|
| aso-marketing | 1.2.1 | max | App Store + Google Play keyword and metadata optimization |
| appstore-review-checker | 1.2.1 | high | Pre-submission audit vs Apple guidelines |
Tooling
| Skill | Version | Effort | What it does |
|---|---|---|---|
| cli-builder | 1.0.5 | high | 5-step CLI tool builder with approval gates |
| ollama-optimizer | 1.1.1 | medium | Hardware-aware Ollama tuning |
| install-script-generator | 2.2.1 | high | Cross-platform install.sh with env detection |
| opencode-runner | 1.4.1 | medium | Delegate work to opencode free cloud models |
| herdr-agent-comms | 1.22.2 | medium | Manage Herdr agent fleets: tile panes, message/wait/read, steer |
| issue-work-loop | 1.3.1 | max | Resolve one GitHub issue via a Herdr implementer→reviewer loop until CLEAN |
| tmux-agent-comms | 2.1.0 | medium | Spawn, message, read CLI agents in tmux |
Suite Folders
Most skills are skills/<name>/. Multi-phase products live under a suite folder: umbrella at skills/<umbrella>/ + phases at skills/<umbrella>/<phase>/.
Current suites: website-cloner (6 sequential phases) and diagram-generator (draw.io + Excalidraw engines behind one router). Install an umbrella or any child. Installers discover both levels.
Mirror the layout for your own multi-skill products.
Project docs
| Doc | Purpose |
|---|---|
| CONTRIBUTING.md | Contribute skills; structure and versioning |
| docs/guide-building-agent-skills.md | Authoring guide (plan → write → validate → distribute) |
| docs/brand_kit.md | Logo files, colors, typography |
| docs/DECISIONS.md | Doc ambiguity resolutions |
| docs/troubleshooting.md | Install/setup validation fixes |
| docs/archive/ | Historical work notes (not current product docs) |
| CHANGELOG.md | Release history |
| SECURITY.md | Vulnerability reporting |
FAQ
Do I need every skill?
No. Pick only what you need. All are independent.
Which agents work?
Any that load external skill files. Installer tools (install.sh:23): Claude Code, Cursor, Windsurf, GitHub Copilot, OpenAI Codex, OpenCode, Google Antigravity.
How do I make my own?
Follow CONTRIBUTING.md, docs/guide-building-agent-skills.md, or patterns from existing skills.
Do skills change my runtime code?
No. They only guide the agent during development.
Get Started
npx skills add https://github.com/luongnv89/skills --skill code-review
All skills · Contribute · MIT
Supported Tool Paths
| Tool | Global | Project |
|---|---|---|
| Claude Code | ~/.claude/skills/<skill>/ | .claude/skills/<skill>/ |
| Cursor | ~/.agents/skills/<skill>/ + .cursor/rules/<skill>.mdc | same |
| Windsurf | ~/.agents/skills/<skill>/ + .windsurf/rules/<skill>.md | same |
| GitHub Copilot | ~/.agents/skills/<skill>/ + .github/instructions/<skill>.instructions.md | same |
| OpenAI Codex | ~/.agents/skills/<skill>/ + ~/.codex/AGENTS.md | same |
| OpenCode | ~/.agents/skills/<skill>/ | same |
| Google Antigravity | ~/.agents/skills/<skill>/ | same |
Project Structure
.
├── skills/
│ └── <name>/
│ ├── SKILL.md
│ ├── scripts/
│ ├── references/
│ └── docs/
└── install.sh / remote-install.sh
# Suite umbrellas also hold child skills:
# skills/<umbrella>/<child>/SKILL.md (install.sh:44-46)
Creating Skills
See CONTRIBUTING.md.
Minimal:
---
name: my-skill
description: "When to use and what it produces. Don't use for X."
license: MIT
effort: medium
metadata:
version: 1.0.0
author: "Your Name"
---
# Agent instructions here. Keep SKILL.md under 500 lines.
Contributing
Read CONTRIBUTING.md and CODE_OF_CONDUCT.md.
Security
See SECURITY.md.
Acknowledgements
- frontend-design draws from Anthropic's patterns.
- Many skills follow the conventions established in the skill-creator lineage.
Website · Claude How-To · Blog