github-issue-creator
关于
This skill automates creating GitHub issues for the MCPSpy project using the `gh` CLI tool. It enforces naming conventions with feat/chore/fix prefixes and maintains proper detail levels. Use it when asked to create issues, report bugs, or document features.
快速安装
Claude Code
推荐/plugin add https://github.com/alex-ilgayev/MCPSpygit clone https://github.com/alex-ilgayev/MCPSpy.git ~/.claude/skills/github-issue-creator在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
GitHub Issue Creator Skill
Automates the creation of well-structured GitHub issues for the MCPSpy project.
Tools and Usage
Use the gh issue CLI tool to create GitHub issues. If the issue body is rather long, write it to a temporary markdown file and use the gh issue create --body-file <file> option.
Issue Naming Convention
- Use standard prefixes:
feat(component):,chore:,fix(component): - Component examples:
library-manager,ebpf,mcp,http,output - Brackets are optional but recommended for clarity
- Keep titles concise and descriptive
Examples
feat(library-manager): add support for container runtime detectionchore: update dependencies to latest versionsfix(ebpf): handle kernel version compatibility issues
Issue Content Guidelines
What to Include
- High-level design notes - focus on the "what" and "why"
- POC-level details - enough to get started, not exhaustive
- Actionable scope - should be implementable by a developer familiar with the codebase
What NOT to Include
- Detailed test plans
- Exhaustive acceptance criteria
- Deep technical specifications
- Code examples (unless absolutely necessary for clarity)
When to Use This Skill
- Creating new feature requests
- Reporting bugs and issues
- Documenting technical debt
- Planning work items for the MCPSpy project
GitHub 仓库
相关推荐技能
content-collections
元Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。
sglang
元SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。
evaluating-llms-harness
测试该Skill通过60+个学术基准测试(如MMLU、GSM8K等)评估大语言模型质量,适用于模型对比、学术研究及训练进度追踪。它支持HuggingFace、vLLM和API接口,被EleutherAI等行业领先机构广泛采用。开发者可通过简单命令行快速对模型进行多任务批量评估。
llamaguard
其他LlamaGuard是Meta推出的7-8B参数内容审核模型,专门用于过滤LLM的输入和输出内容。它能检测六大安全风险类别(暴力/仇恨、性内容、武器、违禁品、自残、犯罪计划),准确率达94-95%。开发者可通过HuggingFace、vLLM或Sagemaker快速部署,并能与NeMo Guardrails集成实现自动化安全防护。
