Gives your AI coding agent ready-made instructions for scaffolding a project, comparing product prices across stores, writing SQLAlchemy database code, and mapping how a codebase fits together. Once installed, the agent applies them on its own when a request matches.
Claude Code
> /plugin marketplace add OmarSaleh506/skills> /plugin install omar-skills@skills
Codex
$ npx skills add OmarSaleh506/skills -a codex
Cursor
$ npx skills add OmarSaleh506/skills -a cursor
Gemini CLI
$ npx skills add OmarSaleh506/skills -a gemini-cli
GitHub Copilot
$ copilot plugin marketplace add OmarSaleh506/skills$ copilot plugin install omar-skills@skills
OpenCode
$ npx skills add OmarSaleh506/skills -a opencode
Windsurf
$ npx skills add OmarSaleh506/skills -a windsurf
Portable, install-once skills for AI coding agents. Each skill is a single
SKILL.md of plain-markdown instructions your agent loads automatically the
moment your request matches — no prompt to paste, no copy-paste drift, the same
discipline every time, in every project.
Copy-pasted prompts rot: you tweak one, forget the rest, and every teammate ends up running a slightly different version. A skill is the prompt — versioned, shared, and installed once — so the agent reaches for it on its own, whether that's Claude Code, Cursor, Codex, Gemini CLI, or Copilot.
Any agent — via the skills installer (Claude Code,
Codex, Cursor, OpenCode + 70 more):
npx skills add OmarSaleh506/skills # pick skills + agents interactively
npx skills add OmarSaleh506/skills -a codex # e.g. install only to Codex
Claude Code — native plugin marketplace (one install pulls in every skill):
/plugin marketplace add OmarSaleh506/skills
/plugin install omar-skills@skills
Manual — each skill is a self-contained folder under skills/<name>/; copy
it into your agent's skills directory, or point the agent at its SKILL.md.
Works with Claude Code, Cursor, Codex, Gemini CLI, and Copilot. Model-invoked skills activate automatically when your request matches their description — you don't have to name them.
| Skill | What it does | Use when | Works in |
|---|---|---|---|
| shop-scout | Comparison-shops a product (or a whole list) across stores: ranks by effective price (after coupons + shipping), judges each seller Trusted / Mixed / Risky, and hunts working coupons. Deterministic output, and it degrades to an honest summary instead of a fake table when it can't read live prices. | You want the cheapest and safest place to buy something online. | Any agent |
| sqlalchemy-patterns | The definitive SQLAlchemy 2.0 / 2.1 reference (async · PostgreSQL), verified against the current docs: 2.0-style models, eager-loading strategy, N+1 elimination, bulk writes and upserts, transactions, async testing, and Alembic — a 30-second cheatsheet and pre-query checklist up front, deep dives loaded on demand, plus 2.1 upgrade notes. | Writing any SQLAlchemy model, query, or migration. | Any agent |
| system-flow-mapper | Reads any codebase (source unchanged) and writes three deliverables to docs/system-flow/: a deep technical map, a jargon-free offline HTML companion for non-technical readers, and an interactive offline HTML reference for engineers. | Onboarding to — or documenting — a codebase's architecture and request flow. | Any agent |
| ai-os-init | Non-destructively scaffolds a five-layer "AI Operating System" into a project: CLAUDE.md, docs/ (architecture + ADRs + runbooks), tools/, and a .claude/ with example skills, hooks, and a docs-auditor agent. Idempotent. | Starting or standardizing a repo for a consistent Claude Code setup. | Claude Code |
A real, unedited run (2026-10-07) on the bundled self-hosted Firecrawl. Watch what it doesn't do:
AJB20 was read off the Amazon listing.
It's marked ❔ unknown because it needs a specific card, so it isn't folded into
the effective price.Without a scraping backend, shop-scout does not invent a table. It returns an honest research summary (real links plus any prices it could actually read) and a one-step guide to enabling Firecrawl. See the skill for the backend tiers.
ai-os-init scaffoldsNon-destructive (adds only what's missing — existing settings.json hooks are merged, never overwritten) and safe to re-run. Preview first with --dry-run:
CLAUDE.md # always-loaded project map & conventions
docs/
architecture.md
decisions/ # ADR log
runbooks/
tools/
scripts/ prompts/
.claude/
skills/ # new-adr, clean-tests
agents/docs-auditor.md # read-only docs-drift auditor
hooks/ # guard-secrets, branch-guard, auto-format,
# typecheck, n+1-guard, audit-log
settings.json # wires the hooks
Two hooks are opinionated by design.
branch-guardblocks force-pushes (--force-with-leaseis allowed) and direct commits/pushes to protected branches (main/master/develop/staging/production);audit-logrecords every bash command (add.claude/command-audit.logto.gitignore— commands can contain secrets). Delete any hook you don't want from.claude/hooks/and remove its entry fromsettings.json. No dependencies beyond Python 3 + bash — the formatter/type-check hooks self-skip when their tool (ruff, prettier, tsc, mypy, …) isn't installed.
Stars on GitHub
Helps you take a change from a rough idea or bug report through to a tested, reviewed result, with each step handled by a focused skill. Its skills cover pinning down domain terms, designing module interfaces, diagnosing hard bugs, and reviewing changes against your repo's coding standards.
@mattpocockA development methodology for coding agents that makes the agent brainstorm before building, follow plans, run parallel work and finish branches in a disciplined way. You get fewer skipped steps and more consistent results from your coding assistant.
@obraGives your coding agent a planning, testing and review routine to follow on every task. It plans before building, checks changes with tests, has a fresh reviewer look for regressions, and saves repeated wins as reusable skills.
@affaan-mDeveloper tools for working on the React codebase itself. Provides skills for extracting error messages, managing and debugging feature flags across release channels, running Flow type checks, and fixing lint and formatting issues before committing.
@facebookGives your AI coding agent ready-made instructions for scaffolding a project, comparing product prices across stores, writing SQLAlchemy database code, and mapping how a codebase fits together. Once installed, the agent applies them on its own when a request matches. Install it with the command on its page. It includes ai-os-init, shop-scout, sqlalchemy-patterns. Its skills do not fire on their own yet. Free and open source.
Is this plugin yours?
Claim it with GitHubSubmit a pluginPromote it