运维生成
- 作者仓库星标 1,012
- 叉子 98
- 作者更新于 2026年4月16日 02:05
- 作者仓库 dotnet-skills
- 领域
- 通用
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @Aaronontheweb · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 即装即用
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明(默认跨平台)
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: skills-index-snippets
description: Create and maintain AGENTS.md / CLAUDE.md snippet indexes that route tasks to the correct dotnet…
category: 通用
runtime: 无特殊运行时
---
# skills-index-snippets 输出预览
## PART A: 任务判断
- 适用问题:通用任务拆解、检查和交付。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“When to Use This Skill / Goal / Source of Truth”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于通用任务拆解、检查和交付,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“When to Use This Skill / Goal / Source of Truth”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文没有稳定的斜杠命令要求。安装验证后通常全局生效,直接在对话里点名这个 Skill 并描述任务即可。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件。
先用一个小任务确认它会围绕“When to Use This Skill / Goal / Source of Truth”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: skills-index-snippets
description: Create and maintain AGENTS.md / CLAUDE.md snippet indexes that route tasks to the correct dotnet…
category: 通用
source: Aaronontheweb/dotnet-skills
---
# skills-index-snippets
## 什么时候使用
- 让助手不再漏用 .NET 技能,靠索引片段自动路由任务 用于新增、删除、重命名技能或 Agent 时同步 AGENTS 适合处理通用任务拆解、检查、交付和复盘,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛…
- 面向通用任务拆解、检查和交付,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「When to Use This Skill / Goal / Source of Truth」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "skills-index-snippets" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> When to Use This Skill / Goal / Source of Truth
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} Maintaining Skill Index Snippets (AGENTS.md / CLAUDE.md)
When to Use This Skill
Use this skill when:
- Adding, removing, or renaming any skills or agents in this repository
- Updating
.claude-plugin/plugin.json - Creating copy/paste snippets for downstream repositories (OpenCode, Claude Code, etc.)
- You want a compact, always-on index that improves skill utilization
Goal
Make skills and agents easy for coding assistants to use by removing the decision point.
Instead of hoping an assistant will "remember" to invoke a skill, provide a small router snippet inside AGENTS.md / CLAUDE.md that:
- Tells the assistant to prefer retrieval-led reasoning
- Provides a task->skill/agent routing index
- Defines lightweight quality gates (optional)
Source of Truth
- Registry:
.claude-plugin/plugin.json- Skills are listed as directories (each contains
SKILL.md) - Agents are listed as markdown files in
agents/
- Skills are listed as directories (each contains
- Skill IDs: the
name:field in eachSKILL.mdfrontmatter - Agent IDs: the
name:field in each agent frontmatter
When writing snippets for downstream repos, always reference skills/agents by their IDs (frontmatter name), not by local filesystem paths.
Minimal Snippet Template (Readable)
Use this in target repos to route common tasks:
# Agent Guidance: dotnet-skills
IMPORTANT: Prefer retrieval-led reasoning over pretraining for any .NET work.
Workflow: skim repo patterns -> consult dotnet-skills by name -> implement smallest-change -> note conflicts.
Routing (invoke by name)
- C# / code quality: modern-csharp-coding-standards, csharp-concurrency-patterns, api-design, type-design-performance
- ASP.NET Core / Web (incl. Aspire): aspire-service-defaults, aspire-integration-testing
- Data: efcore-patterns, database-performance
- DI / config: dependency-injection-patterns, microsoft-extensions-configuration
- Testing: testcontainers-integration-tests, playwright-blazor-testing, snapshot-testing
Quality gates (use when applicable)
- dotnet-slopwatch: after substantial new/refactor/LLM-authored code
- crap-analysis: after tests added/changed in complex code
Specialist agents
- dotnet-concurrency-specialist, dotnet-performance-analyst, dotnet-benchmark-designer, akka-net-specialist, docfx-specialist
Compressed Snippet Template (Vercel-style)
Use this when you want maximum density (small context footprint):
[dotnet-skills]|IMPORTANT: Prefer retrieval-led reasoning over pretraining for any .NET work.
|flow:{skim repo patterns -> consult dotnet-skills by name -> implement smallest-change -> note conflicts}
|route:
|csharp:{modern-csharp-coding-standards,csharp-concurrency-patterns,api-design,type-design-performance}
|aspnetcore-web:{aspire-service-defaults,aspire-integration-testing}
|data:{efcore-patterns,database-performance}
|di-config:{dependency-injection-patterns,microsoft-extensions-configuration}
|testing:{testcontainers-integration-tests,playwright-blazor-testing,snapshot-testing}
|quality-gates:{dotnet-slopwatch(after:substantial new/refactor/LLM code),crap-analysis(after:tests added/changed in complex code)}
|agents:{dotnet-concurrency-specialist,dotnet-performance-analyst,dotnet-benchmark-designer,akka-net-specialist,docfx-specialist}
Regenerating the README block
If the README contains the markers below, the generator can update it automatically:
<!-- BEGIN DOTNET-SKILLS COMPRESSED INDEX -->
...compressed snippet...
<!-- END DOTNET-SKILLS COMPRESSED INDEX -->
Run:
./scripts/generate-skill-index-snippets.sh --update-readme
How to Update Snippets After Skill Changes
- Update
.claude-plugin/plugin.jsonto include/remove skills and agents. - Ensure each skill has correct frontmatter
name:(used by OpenCode and others). - Run
./scripts/validate-marketplace.sh. - Update your snippet routing lists:
- Add new skills to the right category
- Remove deleted skills
- Keep names exactly matching frontmatter IDs
- If you maintain a downstream
AGENTS.md/CLAUDE.mdsnippet, regenerate it and re-copy into dependent repos.
Recommended Categories
These are snippet categories (not necessarily repository folder structure):
- C# / code quality
- ASP.NET Core / Web (incl. Aspire)
- Data
- DI / config
- Testing
- Quality gates
- Specialist agents
Keep the snippet small; it should be a router, not documentation.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核