参考稿二创 Agent Skill：倒推选题、行业迁移、双子代理 DBS 联审与流程凭据校验

Hand a live Claude Code session to OpenAI Codex - conversation, skills, and instructions intact. One command, then Codex opens on your thread. Windows, macOS, Linux. MIT.

Checkpoint-gated academic research skills and agent orchestration from idea to submission.

A battle-tested Claude/Agent Skill for authoring, debugging, packaging and publishing QGIS Python plugins.

Privacy-first Codex Skill for authorized local voice cloning with CosyVoice2, pronunciation control, prosody guidance, and QA—no per-call API fee.

A practical operating guide for working agreements, memory, skills, autonomy boundaries, and verified human-agent collaboration.

Six-Lens Decision Advisor: a hybrid Trader/Lawyer/Developer/Politician/Hacker/Entrepreneur skill for AI coding agents

Evidence-led agent skills for CAD, engineering artifacts, independent review, and bounded revision

Audit and repair a document's reference list end to end. Catches AI-fabricated, misattributed, and misdirected citations -- including identifiers that resolve to an entirely different paper.

Classify every task by engineering level using scope AND ownership, then show how each level would have done the same task. Built on Honeycomb's ladder. Agent skill.

Public-safe Agent Skill for strict parent-child OGSM inheritance and strategy/budget alignment reviews.

Automatically run Chinese university computing labs, capture results, and fill the supplied DOCX template across Windows, Linux, and macOS.

75 production-grade agent skills for Hermes Agent + Paperclip — research, write, organize, earn, and run an AI workforce. Every skill passes a QA gate with hard safety rails. Built by Gen AI Hub.

An agent skill that grades a founder's sales, partnership, or investor call from the transcript. Anchored scores, not encouragement. MIT.

Personal Agent Skills by TyrantLucifer

streamer-mode — a coding-agent skill for live streams and screen shares: never print secrets, env vars, tokens, IPs, or hostnames on screen; safe command alternatives; findings deferred privately.

Source-grounded research to verified HTML artifacts, as a portable agent skill.

轻量 AI 开发工作流：6 个 agent skill + 每项目几个 markdown，零 CLI 依赖、零 hook、零注入税。A lightweight AI dev workflow as agent skills — one hard gate, zero per-project scaffolding.

Open-source Agent Skills for turning AI-assisted drafts into writing that sounds more like you.

Skill do Claude Code para integrar com a API do ERP Conta Azul: OAuth 2.0, 10 armadilhas mapeadas, mapa de ~45 endpoints e referencia completa navegavel.

Entalpa MCP client instructions, Agent Skills, and plugin bundle

Claude Code などの AI エージェント向けセッション運用スキル集（SKILL.md 形式・MIT）

Agent Skill for Schedulala — schedule social posts to 12 platforms from Claude, ChatGPT, Claude Code, or any MCP agent

🧠 Agent skill to prevent AI sycophancy, unverified assumptions & overconfident reasoning

Portable Abbie skills and credential-safe MCP wiring for agent clients.

RTL/Persian UI design skill for AI coding agents (Claude Code, Cursor, Codex) — correctness rules, Persian typography recipes, and a zero-dependency violation detector. Agent Skills standard.

🐘🐟 大象金魚模式 Agent Skill — 四角色分離與 CUJ 驅動開發，附客觀驗證腳本。工具中立，可移植至 Claude Code / Antigravity / Cursor。

Skill mínimo para Codex, inspirado no /recap do Claude Code, que resume a conversa atual em três linhas.

Mirror is a Hermes-native AI as Me workflow for identity, calibration, memory, SkillOS, and EvolutionOS.

Agent Skill for Claude Code & Codex: build, validate, and ship WeChat Mini Programs — AI Mode Skills, CloudBase, privacy scan, layered release SOP, plus a standalone preflight checker. 微信小程序开发、验证与发布 SOP

Author-side manuscript audit skill for pre-submission and revision quality control. Not a substitute for confidential peer review or editorial judgment.

Terminal49 container tracking tools and agent skills for Cursor, Claude Code, Codex, and GitHub Copilot.

Adaptive design direction and execution skill

Cross-platform Agent Skill for safe Jenkins CLI setup, job discovery, build triggering, and monitoring.

Provider-independent agent skill for editorial paper-collage videos

Cross-platform Agent Skill for safe subagent selection, task packets, permissions, handoff, and verification.

Bidirectional Codex and Claude coordination skills and plugins for reviewed, evidence-driven multi-agent delegation.

A customer-support Digital FTE (AI Worker): OpenAI Agents SDK on a sandboxed runtime, capabilities as portable Skills, a scoped customer-data MCP server instead of raw SQL, pgvector semantic search on Neon Postgres, and a transactional audit row behind every action.

Parallel research without context bloat—a Codex and Claude Code skill using read-only subagents, isolated reports, and evidence-backed synthesis.

Public, reusable skills for AI agents.

Free Claude Code Agent Skill: a 30-minute BUILD / RESHAPE / SKIP protocol that validates a product idea with evidence before you build it

AI-powered Agent Skills for designing production-grade database schemas, data pipelines, and data systems

Agent skills for Claude Code and 80+ other AI coding agents — content authoring, dev workflow, and general session helpers. Install via /plugin or npx skills add.

Claude Code skill for calibrated batch production — Ideate/Build/Self-check/Parallelise with a measured, pre-committed trust gate

Coret Agent Kit: one MCP and portable skill for Codex, Claude Code, OpenCode, Pi, OpenClaw, Hermes, and compatible agents

Open-source image & video generation agent skills, powered by ewo's OpenAI-compatible media API. Free to start, pay-as-you-go.

Turn multiple visual references into an attributed, section-level implementation atlas.

Reusable agent skills for resume-aware job screening and controlled recruiter outreach

Free local-first browser release checks for AI-built websites

Claude Code plugin for reproducible AI data analysis. A skill + slash commands that teach Claude Code to build analysis as a cached, reproducible oryxflow pipeline — so AI-written data analysis is trustworthy and reproducible by default.

Conversational weekend family activity finder — runs a SKILL.md agent with live web search, and cites every pick.

Ask GPT-5.6 Pro for a second opinion with your repo files attached. CLI + agent skill for Claude Code and Codex.

Token-efficient AI agent skill that cuts output tokens without breaking grammar. Installs for Claude Code, Cursor, Codex CLI, and Gemini CLI.

A public, agent-agnostic collection of agentskills.io-compliant Claude Code skills, installable via skills.sh, the plugin marketplace, or manual copy.

Create, stress-test, ratify, and evolve a bilateral operating constitution between a person and an AI agent.

Agent Skill and validator for synchronized NXP GUI Guider and generated LVGL workflows

郑钦安三书《医理真传》《医法圆通》《伤寒恒论》原文检索 Agent Skill

NOCK MCP plugin for Cursor, Claude Code, and Codex