Skills for AI agents

An 11-agent investigation pipeline for Claude Code that breaks stuck loops and proves fixes empirically.

My personal agent skills for AI coding assistants

苏州大学本科毕业论文与相关文档 DOCX 生成 Skill | Codex and Claude Code

Public Agent Skills for coding agents

SaaS factory: Next.js starters + spec-kit extensions + Ralph++ autonomous loop. Production-ready in 48h.

Human-in-the-loop Codex skill for safer LinkedIn engagement: finds relevant posts, drafts contextual comments, avoids duplicates, and waits for approval before any action.

Claude Code Plugins

Find stale feature flags safely｜安全清理陈旧 Feature Flags

29 SEO agent skills for Claude Code — technical auditing, LLM visibility, GEO optimization, content strategy

SpecDD skills for GitHub Copilot: orientation, planning, implementation, review, testing, tracing, and documentation workflows for SpecDD projects.

Deterministic Agent Skill: extend or override Hermes runtime behavior via plugins/config/skills/env — without forking the deployed package.

Portable Claude Code dev workflow — skills, commands, agents, security hooks, and a CLAUDE.md template. One-install onboarding; ships zero credentials.

A Claude Code / Agent Skill: stop asserting high-stakes facts from memory — verify against an authoritative source first. Prevents hallucinated figures and fabricated citations.

Audit and port AI agent skills/plugins across agent ecosystems

GenPark skill: Integrate Similarweb Shopper Intelligence and Amazon IQ concepts for e-commerce insights and competitor analysis.

Manages the XMS export automation webhook configuration file (webhook_config.json). Covers field for

Expert AI Agent Skills for SDD, SRE, DevSecOps and Enterprise Engineering

Set of plugins of agents and skills for daily ai work

Local-first AI meeting notes for macOS — record, transcribe, and structure with your own LLM provider.

Agent skills for building production-shaped developer workflows.

Plugin-ready MCP server for capturing visible Windows application windows.

Talk to your Trakt.tv data through any Agent Skills-compatible AI (Hermes, Claude Code, Cursor, …)

Audit and maintain lean AGENTS.md, MEMORY.md, ARCHIVE.md, and repository Markdown setup files for AI coding agents.

A Codex skill for scoped multi-agent delegation, review, and handoff briefs.

Agent workflow skills with sphinx-needs-backed issues, requirements, and ADRs — drop-in replacement for mattpocock/skills

Trace-based quality gates for LLM-assisted work: REQ -> machine-checkable contracts -> ALLOW/DENY with auditable evidence. Python CLI + agent skill (Claude Code / Codex CLI).

Claude Code skill that turns every HTML report your agent generates into something you can answer in place. 17 baseline features wired in a canonical template — textareas after every question, tab-to-fill, POSTED stamp, three-export-button bar, localStorage persistence.

Codex skill pack for SwiftUI, SwiftData, Xcode debugging, app audits, screenshots, accessibility, Liquid Glass, and App Store release workflows.

Codex skill pack for PRD discussion, requirements clarity, PRD writing, audit, engineering handoff, and QA generation

Claude agent skill that generates realistic fictional enterprise data ecosystems — multi-system schemas, populated SQLite, SQL lineage, controlled DQ imperfections, and a generated read-only MCP server.

Codex skill for pragmatic fresh-eyes reviews with Claude Code

帮助 AI Agent 在长期软件项目中记住背景、检查并修复任务结果，并能在不同工具之间接着做。

Turn flaky CI failures into triage reports｜把不稳定 CI 失败变成排查报告

Agent skills that help humans actually understand what the agent is telling them.

A reusable, production-grade .cursor/ engineering package for Cursor / Claude Code / Agent IDEs. Rules, skills, commands, and deterministic safety hooks extracted from a real Agent project.

Evidence-backed code taste reviews for repositories. A Copilot CLI skill plus reviewer subagent that scores simplicity, architectural elegance, usability, and newcomer understandability with cited evidence.

Codex skill for ephemeral delegation to local OpenCode

A structured AI synthesis pass over your second brain. Borrows the shape of a high-dose psychedelic trip as cognitive scaffold. Seven phases. MIT.

Official Claude/Codex/Cursor agent skills for WikiAIG — ground AI agents on structured, human-curated wikis.

Technical preparation toolkit for buy-side roles: portfolio analytics in Python, Claude Agent skills, VBA macros for Excel, and SQL exercises

Paste messy notes. Get better, tighter rewrites that still sound like you.

Express.js 5 + TypeScript REST API starter with layered architecture, strict ESLint/Prettier, Husky git hooks, and AI agent skill files.

DB-ify your migrations — generate a clean post-merge schema snapshot from a folder of SQL migration files. Built for AI agents.

Free GitHub Action and 1 USDC x402 validator for AI-agent wallet spend, deployments, destructive commands, tokens, and credentials

A personal, growing collection of agent skills for coding agents (Claude Code, Codex, ...).

THE definitive curated resource for motion-design-agent-skills | GenCreator 6-Pillar CoE • Starlight Intelligence • Arcanea Creative Execution • Agentic Passive Income Systems

Codex skills for scientific manuscript drafting, peer-review assessment, technical rigor checks, citation auditing, and research writing workflows.

Discipline layer for AI coding agents (Claude Code & Codex): task router + three pipeline modes with run controls, premortem reviews, TDD gates, evidence-based acceptance, and crash-safe resume. Noncommercial license, 138 behavioral evals.

Two TDD-authored Agent Skills for codebase hygiene: unifying-projects (consolidate duplication) and cleaning-up-projects (remove dead code safely)

Codex-native full core port of claude-obsidian: Obsidian wiki memory, source ingest, hybrid retrieval, locks, DragonScale utilities, canvas, lint, fold, and skills.

YLS Codex usage query skill for Codex and Claude Code

Build reusable writing style skills from source material.

A backward-compatible extension to the Agent Skills format: add a flow (stages) and soft/hard rules to a skill, with stage-scoped just-in-time context injection. LLM-guided, not strictly executed. (skill v2, spec-first)

Free mini-eval and scorecard for AI coding agents: Claude Code, Codex, Cursor, Copilot, Windsurf.

分享 skill，也能体面地恰到饭｜一行命令给开源 skill 配打赏入口、记创作分成链、算收益。GitHub Sponsors · 零依赖 · 默认不回传用量