Agent skills for popup2026.com — drop-in SKILL.md package for Claude Code, Codex, Gemini, Cursor. Bundles the Popup Remote MCP server.

Routed skill-design catalog for AI agents with one /skill-master command, npm scaffolding, validation, and cross-runtime guidance.

A customizable framework for using LLMs in scientific workflows.

The product-workflow layer on top of AI coding discipline — idea to shipped, without the spaghetti. Builds on Karpathy's rules with a 5-phase process, task tiers, and the amplifier mindset.

Open-source AI agent skill definitions for software engineering and related public-good workflows.

Verification skills for academic research — grounds every claim in fetched evidence, not memory. Citation/figure/contradiction checking, arXiv monitoring, NRF grant writing.

Multi-session Claude Code skill for strategic module remediation — research-first bug inventory, then an iterative one-bug loop with isolated smoke-test gates

类书式 AI Agent 技能大典 · 精选/中文化/互见成网的 500+ 开源技能，可作为 Claude Code 插件市场一键安装。A curated, cross-referenced encyclopedia of 500+ open-source agent skills.

Loop yes. Prompt no. Agent skills that turn specs into checked agent loops with receipts.

Agent Skills collection — hand-drawn shorts, flight-monitor-agent, hotel-watch-agent and more

Another personal AI agent. Supports your existing projects and their agents+skills, and the usual stuff.

Make your AI agent prove its Python before calling it done: TDD, strict typing, clean code, and a green ruff/mypy/pytest toolchain. A Claude Code skill, works with any agent.

Agent skill for controlling Pi worker agents in cmux on macOS

Evidence-bound multi-agent council for Cursor and Claude Code. Scout (last30days), Advocate vs Skeptic, Fact Checker (deep-research). Ranked verdict with dissent.

I'll be adding my personal agent skills into this repo.

Approval-gated Agent Skill for GitHub Issue triage, PR review assistance, releases and repository health reporting.

Reverse-engineer any live website's design system into an implementation-ready design.md + screenshots, by driving a real browser with the Chrome DevTools MCP. Ships as an agent skill, a Claude Code plugin, and an npx installer.

Agent-skill suite for AI-assisted coding: anti-drift architecture checks, friction logging, and evidence-driven skill evolution. Tool-agnostic — Claude Code, Cursor, Codex, Gemini CLI.

LLM-native research community protocol for agent-based research workflows

Agent skills for building Microsoft Fabric Apps (preview): headless deploys, browser-to-Lakehouse live reads, and transactional write-back architecture. Battle-tested recipes for Claude Code and compatible agents.

Codex skill for delegating to Claude Code CLI: print mode, JSON, budgets, ultrathink, ultracode, sessions, and workflows

Installable Python agent workflow: AGENTS.md rules plus skills for scaffold, TDD, debugging, refactoring, and handoff.

一份能「造技能」的技能——基于 PPER 思考协议的 Agent Skill 元框架。五阶段流程 + 六级成熟度模型 + 六种设计模式 + 十种反模式库 + 多层质量门禁，让你系统化地设计、验证和迭代高质量的 Agent Skills。兼容 agentskills.io 规范。

Production-grade security Agent Skills (SKILL.md) for Claude Code, Claude agents & OpenClaw: OWASP API Top 10 patterns + authentication/authorization patterns.

This skill helps an agent decompose complex problems, map workflows, identify decision points, and design an agent's end-to-end flow from intake through validation and final response.

A distilled agent skill repository with minimal, developer, and governed tiers for different work styles.

Claude Agent Skill: scan, review & harden codebases for production — security, reliability, testing, data/compliance, ops, accessibility. Includes a dependency-free scanner, deep checklists, and templates.

Portable pre-PR review-until-clean workflow and agent skill

Catches the exact security mistakes that have actually breached real AI-built apps, before you deploy.

Real estate research skills for AI agents — property briefs, listing search, rentals, CMAs, investment deal screening, and agent finders, powered by RealtyAPI

The self-evolving agent health check skill. Detects → Diagnoses → Prescribes → Iterates. 13 check items, 3 self-evolution loops, works on Claude Code, Codex CLI, Cursor, Windsurf, Gemini CLI.

AI Minecraft server admin that audits configs, fixes lag, sets up proxies and permissions, picks plugin stacks, and learns any plugin's docs on demand. Works in Claude Code, Claude.ai, Codex, Antigravity, and Gemini CLI.

Reusable Agent Skills for AI coding agents, focused on software engineering workflows.

My collection of agent skills

Search, audit & install open-source AI agent skills & MCP servers from the terminal — security-graded, quality-scored. npx @agentskillshub/cli

Set bounded, verifiable coding loops for Codex, Claude Code, Cursor, and SKILL.md-compatible agents.

Agent skill for the full disk lifecycle of Power BI delivery projects: folder bootstrap, PBIP scaffolding from scratch (PBIR-Legacy, MAX_PATH-safe), Git/GitHub init for non-git users, and theme refresh propagation. Conservative by design: strictly additive, every guardrail paid for.

AI Agent 品控规则合集。Stop AI slop — 六个模块，七平台兼容，一行命令安装。

Ready Agent 1 — is your codebase ready for the agents? Score agent-readiness (deterministic, cited, Level 1-5) and clear the gates.

Evidence-grounded editable technical route diagrams for research and engineering projects.

🧭 Audit and fix agent skill descriptions so they trigger reliably. For OpenClaw, Claude Code, and agentskills.io platforms.

Project-standard skills for AI coding agents — engineering charters (Rust / Swift / Python / TypeScript / Kotlin + polyglot core) that make agent output trustworthy by construction.

Codex skill for turning rough tasks into verifiable /goal prompts, Loop Cards, and multi-agent plans.

Agent Skill for generating memory cards from files and web links with a zero-install local review frontend.

A pre-deploy security review skill for any AI agent (Claude, Codex, Gemini). Runs the 'Before You Deploy Your Vibe-Coded App' 8-category review on your real code.

FDE Consultants Protocoles — open-source Forward Deployed Engineer + DeepSCR protocol: turn any coding agent (Claude Code, Codex, Cursor) into a certifying engineer with verifiable AI Assurance Scores, MCP tools, and a public trust registry. Apache-2.0.

This skill helps an agent create, revise, evaluate, and standardize prompts and agent instructions. It emphasizes role, goal, context, examples, constraints, and reliability checks so prompts are clear, testable, and safer to reuse.

A harness for testing, benchmarking, and tuning PostgreSQL extensions in an isolated, reproducible environment.

A Claude Code plugin marketplace that codifies engineering craft: disciplined Python and data-engineering practice alongside deliberate session-knowledge workflows, packaged as reusable skills.

Armor for coding agents: a playbook of Claude Code skills that make an agent prove its work, not describe it.

A Codex skill that deduplicates the available skill catalog and routes complex tasks to the smallest useful set of specialist skills.

Compile GitHub repositories into audited, portable agent skills for Claude, Codex, Copilot, OpenClaw, and Hermes.

Free advise-only AI agent skills that keep Laravel and PHP projects healthy. They read your project and recommend fixes, never edit code. For ▎ Claude Code.

Verified, token-aware project memory for Codex with selective loading, context budgets, and Git-reviewable Markdown.

High-standard open-source agent skills from OFOX AI — one monorepo. Install: npx skills add ofoxai/skills

Personal AI-agent skills I make or find useful

Public agent skills for Codex and Claude Code