Maintainer of this skill?
by CCharlie-xiu
Task-first AI agent skill library · https://skill.wuch-ai.net
# Add to your Claude Code skills
git clone https://github.com/CCharlie-xiu/aoci-wuchSee how aoci-wuch compares with popular alternatives.
aoci-wuch is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by CCharlie-xiu. Task-first AI agent skill library · https://skill.wuch-ai.net. It has 51 GitHub stars.
aoci-wuch's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/CCharlie-xiu/aoci-wuch" and add it to your Claude Code skills directory (see the Installation section above). aoci-wuch ships a SKILL.md manifest, so compatible agents can discover and load it automatically.
aoci-wuch is primarily written in Python. It is open-source under CCharlie-xiu on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh aoci-wuch against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
本地仓库用于编辑；跨项目复用以 GitHub main 为准。
仓库说明默认英文：README.md；中文见 README.zh.md。站点：https://skill.wuch-ai.net。
https://github.com/CCharlie-xiu/aoci-wuch
不要默认读取能力库。
只有当前任务可能复用已有能力时，才去读：
任务
↓
1. 读能力索引 capabilities/index.md
capabilities-unstable/index.md
↓
2. 找命中的能力 索引一行 = 一个能力
↓
3. 只读该能力 capabilities/<name>/index.md
↓
4. 必要时继续深入 该目录内的其他文件
未命中就按常规方式完成当前任务，不要继续读能力正文。
复用前读取远程索引：
curl -fsSL "https://raw.githubusercontent.com/CCharlie-xiu/aoci-wuch/main/capabilities/index.md"
curl -fsSL "https://raw.githubusercontent.com/CCharlie-xiu/aoci-wuch/main/capabilities-unstable/index.md"
命中后按需读取对应能力文件。常规复用只用正式能力；待验证能力仅在任务适合试用时读取。退役能力只用于追溯或迁移。
本机存在能力库源仓库且其中有 capabilities-pro/ 时（作者本机），另读私有孵化索引 capabilities-pro/capabilities-unstable/index.md；孵化能力仅在任务适合试用时读取，不得把其内容写入公开仓库或公开输出。
远程不可用时回退本地，并说明副本可能过期。远程请求失败不代表索引为空。
本节只负责 Skill Source Self-Update：从 GitHub main 更新当前能力库源仓库。它不扫描、安装或更新 Cursor、Claude Code、Codex、豆包或其他产品中的副本。
请求路由：
执行规则见 follow/update-skills.md。简要流程：
运行 python3 scripts/update_skills.py check。
根据脚本状态处理：UP_TO_DATE 时结束；只有 BEHIND 才展示将要进入的提交和文件差异。
更新前明确告知目标是当前源仓库，并取得用户对本次 fast-forward 的确认。
用户确认后，使用检查结果中的本地与远程 commit 调用 apply：
python3 scripts/update_skills.py apply \
--expected-local <检查时的本地 commit> \
--expected-remote <检查时的远程 commit> \
--confirm
若工作区不干净、分支/上游不匹配、发生分叉或远程不可用，按脚本状态停止；不得 stash、普通 merge、rebase、reset 或覆盖文件来绕过阻断。更新只允许脚本执行 git merge --ff-only。
apply 完成 fast-forward 后自动运行 scripts/validate.py。校验失败时如实报告；不得自动回滚或再尝试其他合并方式。
将“更新 skill”默认理解为更新当前源仓库，但在 apply 前必须展示目标和差异并确认。不要把更新仓库等同于更新已安装副本。
写入前检查工作区，并读取远程规则和目标文件；保留未提交修改。提交并推送后才算发布。
name[tag]: 一句话核心职责
tag = [类型+领域+价值+成熟度+优先级]。完整定义在 _meta/taxonomy.md，需要时再读。
索引行形如 name[tag][PRO]: ... 的是订阅能力，正文不在公开仓库，raw.githubusercontent.com 读不到。
命中 [PRO] 能力时，按顺序：
本机存在能力库源仓库且其中有 capabilities-pro/ 时，直接读取 capabilities-pro/<bucket>/<name>/。
否则用本机读取工具（<file> 缺省为 index.md）：
~/.aoci/fetch capabilities/<name>/<file> # macOS / Linux
powershell -NoProfile -ExecutionPolicy Bypass -File "$HOME\.aoci\fetch.ps1" capabilities/<name>/<file> # Windows
读取工具不存在，或返回未订阅、已过期、本机未激活等提示时，如实转告，给出授权页 https://skill.wuch-ai.net/auth.html，然后按常规方式完成任务。**不得根据索引行猜测或编造正文。**
不要读取或输出 ~/.aoci/ 下的订阅码与工具内容。
follow/ 规则capabilities-pro/capabilities-unstable/，孵化期间不进公开索引、首页与 README（公开历史不可收回）capabilities-pro/，不得出现在公开目录或公开提交中index.html（GitHub Pages）与 README.md，以 scripts/validate.py 通过为准follow/promote.md）只要本次任务读取、引用或使用了能力库内容，回答最后一行必须添加：
【能力库已使用｜本次能力：<能力名称>】
仅读取规则或索引时：
【能力库已使用｜本次能力：规则/索引】
未读取能力库：不添加。
English | 中文
A task-first skill library for AI coding agents.
Reusable workflows, reasoning methods, MCP setup, UI design, mobile, Excel, and video—indexed so the assistant finds a match, then reads only what it needs.
Site → https://skill.wuch-ai.net · Subscribe / auth · Stable index · Experimental index
| You get | Meaning |
|---|---|
| Task → catalog → one skill | No dumping every guide into context at chat start |
| Stable vs Experimental | Maturity is visible; experimental is never treated as proven |
| Free + PRO | Public skills on GitHub; subscription skills via auth |
| Works with Cursor / similar | Install the repo as an Agent Skill, then let the agent follow SKILL.md |
Install this repository as a reusable Agent Skill:
https://github.com/CCharlie-xiu/aoci-wuch.git
Or tell your assistant:
Install https://github.com/CCharlie-xiu/aoci-wuch.git as a reusable Agent Skill for this AI assistant.
User task → Read catalog → Select a match → Load its guide → Do the work
The assistant should not load every capability body up front. Use capabilities/index.md (and the experimental index when appropriate), then open only the hit.
| Capability | Purpose |
|---|---|
| GitHub App Authorization | Run the user authorization flow for a given GitHub App |
Native Mobile App Setup PRO |
Create and maintain iOS / Android apps with the mobile-template family |
| Logic Space Enumeration | Enumerate valid scenarios, paths, and outcomes in complex designs |
| Excel Backend Selection | Choose and wire file-level vs desktop Excel backends |
iOS App Store & TestFlight PRO |
Ship an Xcode app to TestFlight and the App Store |
| DBX Database Setup | Discover, configure, and verify DBX MCP for database access |
| Video Frame Extraction | Extract ordered stills from video from natural-language requests |
| Design Style Canvas | Confirm desired effects against the wuch-ai design anchor, then ship a style Canvas |
UI Style Prompt Design PRO |
Confirm style first, then generate runnable UI |
Lightweight SSO Login Center PRO |
Lightweight SSO Login Center |
| GA4 Analytics | GA4 Analytics |
| Capability | Purpose |
|---|---|
| LibTV Media Setup | Configure and verify LibTV Remote MCP |
| Workers Site Auth & Abuse Control | GitHub OAuth / email login and rate limiting on Cloudflare Workers |
SKILL.md Agent skill entry (read this first after install)
capabilities/ Stable capabilities
capabilities-unstable/ Experimental capabilities
capabilities-retired/ Retired (reference only)
follow/ Write / update / promote / deprecate rules
scripts/ validate.py, release, source self-update
index.html · auth.html Project site (GitHub Pages → skill.wuch-ai.net)
python3 scripts/validate.py
Fast-forward this clone from GitHub main only—does not update copies already installed inside AI products:
python3 scripts/update_skills.py check
python3 scripts/update_skills.py apply \
--expected-local <local-commit-from-check> \
--expected-remote <remote-commit-from-check> \
--confirm
Details: follow/update-skills.md.
All rights reserved. Personal viewing and personal use only; no redistribution of modified copies; external contributions are not accepted. See LICENSE.