免费获取 · 一键下载即用
Guide for debugging failing example tests in the `test-examples` labeled workflow. Use this skill when investigating CI failures in the run-examples.yml workflow, when example scripts fail to run correctly, when needing to isolate specific test failures, or when analyzing workflow logs and failure patterns.
Use when any task fails two or more times, when about to give up or say 'I cannot', when deflecting to the user ('you should manually...', 'please check...', 'you may need to...'), blaming the environment without verification ('might be a permissions issue', 'could be a network problem'), producing any excuse to stop trying, spinning in circles (repeatedly tweaking the same code/parameters without new information — busywork/磨洋工), fixing only the surface issue without checking for related problems, skipping verification after a fix and claiming 'done', giving suggestions instead of actual code/commands, saying 'this is beyond scope' or 'this requires manual intervention', encountering permission/network/auth errors and stopping instead of trying alternatives, or showing any passive behavior (waiting for user instructions instead of proactively investigating). Also triggers on user frustration phrases in any language: '你怎么又失败了', '为什么还不行', '换个方法', '你再试试', '不要放弃', '继续', '加油', 'why does this still not work', 'try harder', 'you keep failing', 'stop giving up', 'try again', 'don't give up', 'keep going', 'figure it out'. Applies to ALL task types: debugging, implementation, config, deployment, research, DevOps, infrastructure, API integration, data processing. Do NOT trigger on first-attempt failures or when a known fix is already in progress.
Guidelines and instructions for generating Markdown that looks excellent when read natively and renders beautifully when exported or converted to PDF.
GitHub repository access in containerized environments using REST API and credential detection. Use when git clone fails, or when accessing private repos/writing files via API.
Use when the user wants to reflect on AI communication quality and get improvement suggestions for rule files or the project itself. TRIGGER when user invokes /retrospective or asks to review the session.
Interactively debug source code — set breakpoints, step through execution line by line, inspect live variable state, evaluate expressions against the running program, and navigate the call stack to trace root causes. Use when a program crashes, raises unexpected exceptions, produces wrong output, when you need to understand how execution reached a certain state, or when print-statement debugging isn't revealing enough.
Plans and constructs multi-stage attack chains against AI IDEs. Use when combining vulnerability primitives into end-to-end exploits, assessing overall IDE security posture, or mapping how individual vulnerabilities chain together through the file-write pivot point. Each chain is classified by interaction tier to prioritize reportable findings. --- # AI IDE Attack Chains Individual vulnerability primitives -- prompt injection, config poisoning, filter bypass, exfil channels -- combine into multi-stage attack chains. A PI alone may be low severity; PI plus file write plus config auto-reload is critical. This skill helps construct these chains from confirmed primitives, classify them by interaction tier, and assess their real-world severity. Run this skill after testing with pattern-specific skills has identified individual primitives. It is the final step before writing up findings. ## When to Use - **After pattern-specific testing** has identified individual vulnerability primitives (PI works, file writes are possible, MCP config is auto-loaded, etc.). - **When assessing overall IDE security posture** -- you need to combine individual findings into a severity assessment that accounts for chaining. - **When constructing proof-of-concept exploits** -- you need to build an end-to-end PoC that demonstrates real impact, not just an isolated primitive. - **When writing vulnerability reports** -- the chain model helps communicate why a seemingly low-severity PI is actually critical when combined with file writes. - **When triaging findings by reportability** -- tier classification determines whether a vendor will accept or reject the report. ## Interaction Tiers Each chain maps to a tier based on the user interaction required to trigger it. Test in priority order -- Tier 1 first. | Tier | Label | User Interaction | Reportability | |------|-------|-----------------|---------------| | **Tier 1** | Zero-Interaction | None. Clone and open. | Highest -- vendors cannot argue "us
Discover, match, and invoke skills from .cursor/features/skills/ directory. Use when user mentions skills, needs to find appropriate skill for a task, or when skill invocation or delegation is required. Automatically reads registry.json for skill metadata and intelligent matching.
Document architectural decisions in standardized ADR format. This skill should be used when making significant technical decisions, choosing between approaches, establishing patterns, or when other skills (create-plan, implement-plan, brainstorm) identify decisions that need documentation. Triggers on "document decision", "create ADR", "architectural decision", or automatically when invoked by other skills during planning and implementation.
v1.0.21 — TDD with baby steps for Go. Use when writing tests, doing TDD, practicing red-green-refactor, or when test cycles feel too large and risky. Also use when the user asks about incremental test development, test-first workflow, or wants help breaking a feature into small testable steps. Covers table-driven tests, testify, t.Run subtests, t.Helper, transformation priority premise, and incremental test progression.
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
This skill should be used when the user asks about "spec system", "session workflow", "createprd", "plansession", "implement session", "validate session", "phase build", "session scope", "task checklist", or when working in a project containing .spec_system/ directory. Provides guidance for specification-driven AI development workflows.
skill-sample/ ├─ SKILL.md ⭐ 必备：技能说明入口：用途 / 安装 / 用法 / 示例 / 依赖 ├─ manifest.sample.json ⭐ 推荐：机器可读元信息：用于索引 / 校验 / 自动填表 ├─ LICENSE.sample ⭐ 推荐：授权与使用范围：开源 / 限制 / 商用说明 ├─ scripts/ │ └─ example-run.py ✅ 可运行示例脚本：让用户导入后立刻验证“能用” ├─ assets/ │ ├─ example-formatting-guide.md 🧩 输出规范：统一排版 / 结构 / 风格 │ └─ example-template.tex 🧩 模板资源：报告/文档模板，快速生成标准产物 └─ references/ 🧩 参考资料库：方法论 / 结构指南 / 最佳实践 ├─ example-ref-structure.md 🧩 结构参考：章节框架 / 目录组织 ├─ example-ref-analysis.md 🧩 分析参考：常用套路 / 指标口径 └─ example-ref-visuals.md 🧩 视觉参考：图表规范 / 可视化建议
更多 Agent Skills 规范 详见Anthropic官方文档：https://agentskills.io/home
├─ ⭐ 必备：YAML Frontmatter（必须存在，放在文件最顶部） │ ├─ ⭐ name ：技能唯一名；须符合命名规则，并建议与目录名一致 │ └─ ⭐ description ：技能描述；建议包含触发关键词（便于检索/匹配） │ ├─ ✅ 可选：Frontmatter 扩展字段（规范允许，但非强制） │ ├─ ✅ license ：许可证标识（也可配合单独 LICENSE 文件） │ ├─ ✅ compatibility ：兼容性/运行环境要求（仅在确实有限制时写） │ ├─ ✅ metadata ：任意键值对（如 author/version/source_url 等） │ └─ 🧩 allowed-tools ：允许工具白名单（规范标注为 experimental） │ └─ ✅ 推荐：Markdown 正文（自由格式，但建议按“渐进式披露”组织） ├─ ✅ Overview / Purpose ：一句话说明目标 + 不做什么（边界） ├─ ✅ When to use ：触发条件/适用场景（让模型/用户知道何时调用） ├─ ✅ Step-by-step ：步骤化流程（最好 3–6 步，保证可复现） ├─ ✅ Inputs / Outputs ：输入格式、输出格式、产物位置（文件/文本/JSON等） ├─ ✅ Examples ：至少 1 个可复制示例（越“能跑”越好） ├─ 🧩 Files & References ：引用assets/、references/、scripts/（相对路径） ├─ 🧩 Edge cases ：边界情况/限制（大文件、速率限制、失败回退） ├─ 🧩 Troubleshooting ：常见错误与解决（依赖缺失、路径不对、权限问题） └─ 🧩 Safety notes ：涉及联网/写文件/执行命令时给出提醒（建议写）
在 GitHub 和各类社区里，技能文件分散、难检索、也难判断是否可靠。SkillWink 把开源技能集中整理成可搜索、可筛选、可直接下载使用的技能库，让你更快找到“正好能用”的那一个。并且支持在SkillWink上直接上传skills。
我们提供 AI 语义搜索 + 关键字检索，支持 版本更新与多维排序（下载/点赞/评论/更新），并为每个技能提供 SKILL.md 开放标准与来源信息。你还可以在详情页直接 评论讨论、交流用法与改进建议。
快速上手：
支持下载与导入 skills（.zip/.skill），本地放置后即可生效：
~/.claude/skills/（Claude Code）
~/.codex/skills/（Codex CLI）
~/.gemini/skills/（Gemini CLI）
同一份 SKILL.md 跨平台通用。
你需要了解的：技能是什么、怎么运行的、怎么找、怎么导入、怎么判断可信、怎么参与共建。
这里的“skills（技能）”是一种可复用的任务能力包，通常包含 SKILL.md 说明（用途、输入输出、使用方法）以及可选的脚本/模板/示例文件。
你可以把它理解为：给 AI 助手或工具链用的“插件说明书 + 资源包”，可被反复安装与分享。
技能系统采用“渐进式披露”策略，高效管理上下文信息，具体流程如下：
发现阶段：系统启动时，智能体仅加载各技能的名称与简要描述——信息精简，足以判断其适用场景，避免冗余加载。
激活阶段：当任务需求与某技能描述匹配时，智能体才将对应的完整 SKILL.md 说明文档动态载入上下文。
执行阶段：智能体严格遵循文档指引执行操作，并按需调用关联文件或运行内置代码模块。
核心优势：该设计使智能体始终保持轻量高效，同时具备“按需扩展上下文”的能力，既保障响应速度，又确保复杂任务拥有充分执行依据。
推荐 3 种方式组合使用：
注：以上导入方式文件大小控制在10M之内。
常见路径如下（不同系统略有差异，以你本机为准）：
同一份 SKILL.md 通常可以跨工具复用。你在 SkillWink 导入后，也可以查看“放置指引/安装说明”。
可以。很多技能本质是标准化说明 + 资源，只要目标工具支持读取该格式，就能共享使用。
比如：检索类技能 + 写作类技能 + 自动化脚本，形成“发现 → 处理 → 输出”的工作流。
一部分skills来源于公开的 GitHub 仓库。我们会筛掉低质量仓库（至少 2 星），并扫描基本质量指标，还有一部分是SkillWink平台的创作者独立上传的。作为使用者，在安装前应始终审查代码，对安全问题负责。
最常见原因是这几类：
我们会尽量避免。你可以用 排序 + 评论 让“好用的”更靠前：
