Art & Design

我的 Codex CLI Skill 合集：AI PM 全流程、品牌 DESIGN.md 风格库、小红书发布、玄学用户画像

Per-task isolated worktrees for parallel Apple development with AI agents. Run multiple Claude / Codex / Copilot / Cursor sessions on the same Xcode project without build.db locks, DerivedData thrash, or simulator collisions.

A terminal-native coding agent harness — model-agnostic, hackable guides, sensors, hooks, skills and tools for agentic development.

Forkable system for generating LinkedIn content with Claude — the 60-min workshop, the skills, the VOICE.md persona, and the humanizer pass. Built around Ilija Cosic's actual workflow at Expandi.io.

Image generation skill for skill-host agents (text2im / image edit / transparent / sprites). Fork of OpenAI Codex CLI sample skill (Apache 2.0).

Self-improving AI pipeline that autonomously discovers, builds, audits, and ships improvements to any codebase. 4 agents, 5 phases, eval gating, continuous learning. Plugin for Claude Code & Gemini CLI.

Ottomate — your universal AI agent workbench. Give it a goal — it plans, codes, browses, and delivers autonomously. 190+ connectors, advanced multimedia creative suite, browser automation, visual pipelines, 200+ skills, cron task scheduler, and more

Turn Hermes into a local-first wellness agent with guided setup, onboarding, wellness skills and MCP connector presets.

An AI orchestration workbench for building visual flows, agent-driven skills, and OpenAI-compatible plugin workflows.

An autonomous dev-loop skill for Claude Code: /devloop works a queue of issues one at a time — branch → implement → build & test → PR → sub-agent review → merge → verify the deploy → next. Self-paced; runs unattended.

Elevate your skills with hands-on training in GenAI and AI Agents at Sathyabama University. Join us to build real-world AI solutions! 🚀💻

MMSkills: Towards Multimodal Skills for General Visual Agents

Skill for Claude Code / OpenAI Codex: audit a reference site and produce a differentiated parity plan (UI, interactions, API, data, architecture).

Generic OpenAI Responses API agent runtime with dev debugging UI, public chat UI, skills, MCP diagnostics, and tool-call tracing.

AI agent core designed to build Vertical AI Agents. Self-improving. Built-in learning loop - it creates skills, scripts from experience, improves them during use.

Reusable CI workflow for QuickClaw-Skills org — validates SKILL.md, skill.json, forbidden files, secrets, paths, syntax, release sync 3-way.

Reusable skills for document parsing and agent workflows, turning PDFs, DOCX, PPTs, and images into LLM-ready Markdown.

Claude Code skills for automating qualitative UX research - transform interview transcripts into personas, journeys, and strategic insights

Local control panel for Claude Code — plugins, skills, hooks, sessions, memory in one UI. Edit CLAUDE.md and add hooks back to settings.json. Pure stdlib.

基于 Claude Code Skills 的测试用例自动生成管道：飞书 PRD → 五维评审 → 飞书测试用例文档

Starter repository template for my projects. Includes style guides, AGENTS.md, agent skills, and a small set of reusable development scripts to bootstrap new repos with consistent structure and workflow.

VeriFlow-CC: A Claude Code-driven RTL design pipeline. Automates Chip-on-Chat from architecture to synthesis (iVerilog/Yosys) using a stateful, zero-dependency LLM orchestration skill. Features sub-agent nesting for code gen and behavioral-driven verification.

The official Thin Runner Sandbox: a practical learning environment for building transparent AI workflows without heavyweight black-box agents. Includes learning modules, skills, pipelines, memory systems, observability examples, Ollama/OpenAI integrations, and production-ready orchestration patterns.

A collection of skills I made for AI agents and models to speed up my workflow as well as yours