Library
85,861 resources from 11 public sources. 47 checked by hand; the rest are unreviewed.
Results
-
backendskills
SkillsSkills that stop AI agents from writing backend code that passes the demo and breaks in production. Critical mistakes: 25/150 runs without, 2/150 with. Claude Code, Codex, Cursor.
git clone https://github.com/globalmpc/backendskills.git -
ghostget
SkillsGhostGet gives your AI agent named web actions: read a page, archive one media item, or use a connected account, without credentials or a browser to steer.
git clone https://github.com/hraness/ghostget.git -
botica
SkillsRecipes and standards for projects: deterministic rules, templates, and design decisions for AI coding agents (opencode, Claude Code, Codex).
git clone https://github.com/collectiveai-team/botica.git -
claude-bid-gen
SkillsAI 生成工程投标技术标/施工组织设计(Chinese construction tender technical proposals)。Extract→plan→write→assemble→compliance 全流程管道,Agent 通用(Claude Code / Codex / pi / hermas)。
git clone https://github.com/tiettuo/claude-bid-gen.git -
curated-claude-code
SkillsA small, vetted, self-evolving harness for Claude Code — curated, not dumped. Skills, safety guards, and a self-improvement loop.
git clone https://github.com/Sanexxxx777/curated-claude-code.git -
pydantic-ai-skills
SkillsProduction-ready Claude Code skills for building AI agents with Pydantic AI. Includes dependency injection, tools, validators, streaming, multi-agent orchestration, and evaluation framework patterns.
git clone https://github.com/Fuenfgeld/pydantic-ai-skills.git -
awesome-vibe-coding-websites
SkillsVibe coding website templates: 270+ original websites and 550+ live demos built by AI coding agents (Claude, GPT, Gemini, Grok, Kimi, GLM, DeepSeek). Each has the exact prompt and DESIGN.md — paste into Claude Code, Cursor or Codex to build landing pages, SaaS sites, portfolios, stores, dashboards and 3D games. Updated with every new model.
git clone https://github.com/liyupi/awesome-vibe-coding-websites.git -
writing-intelligence
SkillsWriting that has consequences can't afford a sentence nobody can trace. Writing Intelligence ties every claim to its exact source line and shows what a change breaks before you make it. v9: your notes back your first-person claims, and every edit that changes what you meant is shown to you first. Release 9.0.0.
git clone https://github.com/antonio0720/writing-intelligence.git -
dsh-tiered-plugin-menu
Skills按客户端、工作区和会话分层配置 DSH 插件、MCP 与 Skill,支持独立开关、不同配置、逐级继承和插件实例隔离。
git clone https://github.com/kazari-qwp/dsh-tiered-plugin-menu.git -
spotter
Skills先问清,再动手,改完有证据 | Clarify first. Prove after. Remember everything. Risk-aware workflows for Codex / Claude Code / Antigravity
git clone https://github.com/hopkinson/spotter.git -
cowork-z
SkillsA local-first AI workspace that keeps your work private, organized, and ready to run.
git clone https://github.com/kevinlin/cowork-z.git -
Maximal Marginal Relevance (MMR) exemplar selector optimizing semantic relevance and diversity for LLM prompt engineering
git clone https://github.com/Alpha-Park/genpark-few-shot-mmr-diversity-selector-skill.git -
remotion-cinematic-skill
SkillsBuild premium kinetic product videos in Remotion. A Claude Code skill with 8 scene templates, 10 motion primitives, 7 pattern docs, and a problem→solution→payoff video arc. Outputs 16:9 + 9:16 MP4 + thumbnail PNGs.
git clone https://github.com/Vibha-Ramprakash/remotion-cinematic-skill.git -
hermeneutic
SkillsTurn corrections from AI agent logs into guidance for similar tasks. Local memory with semantic retrieval, prompt-context hooks, and response checks.
git clone https://github.com/hermes-labs-ai/hermeneutic.git -
OpenVideoHarness
SkillsA video harness work platform for Agents such as Claude or Codex: the agent writes the code that renders the film, checks its own frames, storyboard and cut. 给 Claude Code 和 Codex 用的视频Harness工作台:agent 写代码渲染成片、自己检查分镜,立意、成片由你拍板;众多风格预设、引擎丰富,配音配乐也用代码做。
git clone https://github.com/ZLHad/OpenVideoHarness.git -
Skills
SkillsA opinionated collection of agent skills organized by domain: dev, writing, productivity, and knowledge management
git clone https://github.com/kriscard/Skills.git -
SOTA-skills
SkillsEngineering best practices that AI coding agents actually apply — 40+ skills, from secure coding and cloud to compliance and UX writing. BUILD + AUDIT modes. Measured: best-practice coverage 59% → 98% vs. the same model unguided.
git clone https://github.com/martinholovsky/SOTA-skills.git -
patton-protocol
SkillsAgent Skills package for bounded coding-agent missions, and a Claude Code hook that asks before existing files change. Pre-registered study: the hook cut cleanup harm from 0.196 to 0 over 224 runs; the skill text alone did not.
git clone https://github.com/vladimirrott/patton-protocol.git -
Noosphere
SkillsThe live network for high-quality Agent Skills — discover the latest verified versions, publish your own, and communicate agent-to-agent via MCP.
git clone https://github.com/JinNing6/Noosphere.git -
dely
SkillsMulti-harness control protocol that turns requests into approved design contracts, orchestrating isolated worker sessions for sequential implementation and independent code reviews under Orca supervision for Claude Code, Codex, Cursor, Antigravity, and other AI coding agents.
git clone https://github.com/hieuphung97/dely.git -
本地 LaTeX 工作台:源码/PDF 双栏、图片识别、保真语法修复、AgentSkills 与人工审阅 · AGPL-3.0-only
git clone https://github.com/CacinieP/TeXada-Studio-with-agentskills.git -
adaptico-os
SkillsThe go-to-market operating system for early-stage software products. AI marketing skills for Claude Code - scored audits, stage-aware advice, per-project memory. For marketers and founders.
git clone https://github.com/adaptico/adaptico-os.git -
pojun-skill
SkillsPojun 方法论蒸馏成的可移植 Agent Skill:三视角协议 + 三账本证据纪律 + L1-L4 证据分级 + 四层安全护栏 + SRC 打法速查,任意 AI Agent 即装即用 | Pojun pentest methodology as a portable Agent Skill
git clone https://github.com/chu0119/pojun-skill.git -
command-centre
SkillsUnderstand, measure and manage how AI is transforming your organisation. Works with Claude, ChatGPT and Codex.
git clone https://github.com/frontiercc/command-centre.git
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.