Token导航 LogoToken导航

Agent Skill · 研究检索

oss(操作系统)

oss 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

oss 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为执行命令;许可证为 MIT,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 69,749。

总安装 216周安装 9GitHub Stars 公开资料未说明下载 72核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:oss(操作系统)
来源仓库:https://github.com/paulrberg/dot-claude
仓库路径:skills/oss
安装命令:
npx skills add paulrberg/dot-claude --skill "oss"
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

AgentSkills.tonpx skills
npx skills add paulrberg/dot-claude --skill "oss"

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

oss 用于发现和安装 AI 代理的技能。适用于 Codex、Claude、Cursor、Gemini CLI 中快速集成新工具的场景。通过 npx 和 GitHub 仓库路径安装,支持技能管理与复用。使用前需确认权限范围、维护状态,注意可能触发的网络请求与代码下载。建议结合官方文档验证技能兼容性与实际用途。

来源仓库内路径:skills/oss

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add paulrberg/dot-claude --skill "oss";命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

oss 怎么安装?

在终端执行 npx skills add paulrberg/dot-claude --skill "oss"。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

oss 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

oss 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 npx skills add paulrberg/dot-claude --skill "oss" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

oss 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

oss 用的人多吗?

公开数据是累计安装 216 次,近一周安装 9 次,下载 72 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

oss 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 要求有安全扫描结论才肯安装的团队。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
oss

操作系统 · 本页

216—执行命令MIT未收录

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

OSS Contribution Workflows

Overview

Facilitate GitHub-based open source contribution workflows including pull requests, issues, and discussions. This skill provides structured automation for common OSS contribution tasks, ensuring consistent formatting, proper metadata application, and adherence to repository conventions. Use this skill when contributing to open source projects or managing GitHub repositories through the command line.

The skill emphasizes semantic analysis over mechanical operations—understand the intent and context of changes before generating titles, descriptions, or selecting templates. All generated content should be conversational and informal rather than robotic or overly formal.

Prerequisites

Before using any OSS contribution workflow, verify GitHub CLI authentication status:

gh auth status

Ensure the output shows you are logged in to github.com with appropriate scopes (repo, read:org, etc.). If not authenticated, run:

gh auth login

For pull request workflows, also verify:

  • Working tree is clean or changes are committed
  • Current branch has commits ahead of the base branch
  • Remote tracking is configured for the current branch

Related Skills

For detailed GitHub CLI command syntax, flags, and patterns, activate the gh-cli skill. That skill covers the full gh command surface including issue management, PR operations, repository queries, workflow triggers, and API interactions.

Pull Requests

Create GitHub pull requests with semantic change analysis and structured formatting. The workflow generates meaningful titles and descriptions by reading actual code changes rather than relying solely on commit messages or filenames.

Workflow Steps

Validate Prerequisites

Confirm git state before proceeding:

  • Check git status for uncommitted changes
  • Verify branch has commits ahead of base branch via git log
  • Confirm remote tracking is configured
  • Ensure gh CLI is authenticated

Parse Arguments

Support the following arguments:

  • base-branch: Target branch for the PR (default: main/master)
  • reviewers: Comma-separated list of GitHub usernames
  • --draft: Create as draft PR
  • --test-plan: Include test plan section in body

Semantic Change Analysis

Read the actual diff to understand what changed:

git diff base-branch...HEAD

Analyze:

  • Nature of changes (feature, fix, refactor, docs, test, chore)
  • Scope of impact (which components/modules affected)
  • Intent and purpose (why the change was made)
  • Dependencies or related changes

Generate Title and Body

Craft a concise title summarizing the change. Generate a body with:

  • Summary section (1-3 bullet points explaining what and why)
  • Test Plan section if --test-plan flag provided (bulleted checklist)
  • GitHub admonitions (NOTE, TIP, IMPORTANT, WARNING, CAUTION) where appropriate

Pass the body using HEREDOC syntax:

gh pr create --title "the pr title" --body "$(cat <<'EOF'
## Summary
- First bullet point
- Second bullet point

## Test Plan
- [ ] Test item one
- [ ] Test item two
EOF
)"

Create Pull Request

Execute the gh command with all parsed flags and generated content. Include:

  • --base for target branch
  • --reviewer for each reviewer
  • --draft if draft mode requested
  • Return the PR URL to the user

For the complete pull request workflow with detailed steps, examples, and edge cases, refer to ./references/pull-requests.md.

Issues

Create GitHub issues with automatic template selection and label application. The workflow intelligently chooses the appropriate issue template based on the description content and applies relevant labels for repositories where you have write access.

Workflow Steps

Check for Issue Templates

Search the repository for issue templates:

ls -la .github/ISSUE_TEMPLATE/

Templates may be in YAML or Markdown format. Parse YAML frontmatter to extract:

  • Template name
  • Description
  • Labels to auto-apply

Auto-Select Template

Analyze the issue description to determine the most appropriate template:

  • Bug reports: errors, crashes, unexpected behavior
  • Feature requests: enhancements, new functionality
  • Documentation: docs improvements, clarifications
  • Questions: how-to, usage inquiries

If no template matches or templates don't exist, proceed without one.

Apply Labels

For repositories where you have write access, apply labels using the --label flag:

gh issue create --label "bug" --label "needs-triage"

Extract labels from:

  • Selected template metadata
  • Issue description content (infer appropriate labels)

Support Similar Issue Search

When --check flag is provided, search for similar existing issues before creating:

gh issue list --search "search terms" --state all

Display results and ask user if they want to proceed with creation.

Create Issue

Execute the gh command with generated title, body, and metadata. Pass multi-line bodies using HEREDOC syntax. Return the issue URL to the user.

For the complete issue creation workflow with template examples, label strategies, and search patterns, refer to ./references/issues.md.

Claude Code Issues

Create issues specifically in the anthropics/claude-code repository with environment information gathering and specialized templates. This workflow is optimized for reporting bugs, requesting features, or improving documentation for Claude Code itself.

Workflow Steps

Identify Issue Type

Determine which template to use based on user intent:

  • bug_report: Errors, crashes, unexpected behavior
  • feature_request: New functionality or enhancements
  • documentation: Docs improvements or clarifications
  • model_behavior: Issues related to Claude's responses or reasoning

Gather Environment Information

Collect relevant environment details:

# Claude Code version
claude --version

# Operating system
uname -a

# Terminal emulator (macOS)
echo $TERM_PROGRAM

# Shell
echo $SHELL

Generate Issue Body

Create structured content following the selected template format:

  • Clear description of the issue or request
  • Steps to reproduce (for bugs)
  • Expected vs actual behavior (for bugs)
  • Environment information section
  • Additional context or screenshots

Use GitHub admonitions for important callouts:

> [!IMPORTANT]
> This issue affects all macOS users on Apple Silicon

Create in anthropics/claude-code

Execute the gh command targeting the correct repository:

gh issue create \
  --repo anthropics/claude-code \
  --title "Issue title" \
  --body "$(cat <<'EOF'
Issue content here
EOF
)"

Return the issue URL to the user.

For the complete Claude Code issue workflow with template examples and environment gathering scripts, refer to ./references/issues-claude-code.md.

Codex CLI Issues

Create issues specifically in the openai/codex repository with environment information gathering and specialized templates. This workflow is optimized for reporting bugs, requesting features, improving documentation, or reporting VS Code extension issues for Codex CLI.

Workflow Steps

Identify Issue Type

Determine which template to use based on user intent:

  • 2-bug-report: Errors, crashes, unexpected behavior in CLI
  • 4-feature-request: New functionality or enhancements
  • 3-docs-issue: Docs improvements or clarifications
  • 5-vs-code-extension: Issues with the VS Code/Cursor/Windsurf extension

Gather Environment Information

Collect relevant environment details:

# Codex CLI version
codex --version

# Platform (macOS/Linux)
uname -mprs

# For VS Code extension issues
code --version

Generate Issue Body

Create structured content following the selected template format:

  • Clear description of the issue or request
  • Steps to reproduce (for bugs)
  • Expected vs actual behavior (for bugs)
  • Environment information section (version, platform, model, subscription)
  • Additional context

Create in openai/codex

Execute the gh command targeting the correct repository:

gh issue create \
  --repo openai/codex \
  --title "Issue title" \
  --body "$(cat <<'EOF'
Issue content here
EOF
)"

Return the issue URL to the user.

For the complete Codex CLI issue workflow with template examples and environment gathering scripts, refer to ./references/issues-codex-cli.md.

Discussions

Create GitHub discussions using the GraphQL API with automatic category selection. Discussions are ideal for Q&A, ideas, announcements, and general community conversations that don't fit the issue format.

Workflow Steps

Fetch Repository Information

Query the repository to get its node ID:

gh api graphql -f query='
  query($owner: String!, $repo: String!) {
    repository(owner: $owner, name: $repo) {
      id
    }
  }
' -f owner="OWNER" -f repo="REPO"

Fetch Discussion Categories

Query available categories for the repository:

gh api graphql -f query='
  query($owner: String!, $repo: String!) {
    repository(owner: $owner, name: $repo) {
      discussionCategories(first: 10) {
        nodes {
          id
          name
          description
        }
      }
    }
  }
' -f owner="OWNER" -f repo="REPO"

Auto-Select Category

Analyze the discussion title and body to choose the most appropriate category:

  • Ideas: Feature suggestions, proposals, brainstorming
  • Q&A: Questions seeking answers, how-to inquiries
  • General: Broad discussions, community topics
  • Announcements: Important updates (if you have permissions)
  • Show and Tell: Demos, showcases, sharing work

If uncertain, default to "General" or ask the user to select.

Support Similar Discussion Search

When --check flag is provided, search for similar existing discussions:

gh api graphql -f query='
  query($owner: String!, $repo: String!, $search: String!) {
    repository(owner: $owner, name: $repo) {
      discussions(first: 5, orderBy: {field: CREATED_AT, direction: DESC}, filterBy: {query: $search}) {
        nodes {
          title
          url
        }
      }
    }
  }
' -f owner="OWNER" -f repo="REPO" -f search="search terms"

Display results and confirm before proceeding.

Create Discussion

Execute the GraphQL mutation to create the discussion:

gh api graphql -f query='
  mutation($repositoryId: ID!, $categoryId: ID!, $title: String!, $body: String!) {
    createDiscussion(input: {
      repositoryId: $repositoryId
      categoryId: $categoryId
      title: $title
      body: $body
    }) {
      discussion {
        url
      }
    }
  }
' -f repositoryId="REPO_ID" -f categoryId="CATEGORY_ID" -f title="Title" -f body="Body"

Return the discussion URL to the user.

For the complete discussion workflow with GraphQL examples, category selection logic, and search patterns, refer to ./references/discussions.md.

Common Patterns

Shared conventions and patterns used across all OSS contribution workflows.

HEREDOC for Multi-line Content

Always use HEREDOC syntax when passing multi-line bodies to gh commands:

gh pr create --title "Title" --body "$(cat <<'EOF'
First paragraph

Second paragraph
EOF
)"

The single quotes around 'EOF' prevent variable expansion, ensuring literal content is passed.

GitHub Admonitions

Use GitHub-flavored markdown admonitions to highlight important information:

> [!NOTE]
> Useful information that users should know

> [!TIP]
> Helpful advice for doing things better

> [!IMPORTANT]
> Key information users need to know

> [!WARNING]
> Urgent info that needs immediate attention

> [!CAUTION]
> Advises about risks or negative outcomes

Apply these judiciously—overuse reduces their impact.

Semantic Analysis Over Mechanical Operations

Never generate content based solely on filenames or commit messages. Always read the actual changes:

# Read the diff
git diff base-branch...HEAD

# Or for specific files
git diff base-branch...HEAD -- path/to/file

Understand:

  • What changed (the mechanics)
  • Why it changed (the intent)
  • How it fits into the broader system (the context)

Use this understanding to craft meaningful titles and descriptions that communicate value, not just mechanics.

Informal Tone

Generated content should be conversational and approachable:

Good:

This PR adds support for parsing YAML frontmatter in issue templates. Previously, we only supported markdown format, which meant users couldn't take advantage of GitHub's newer template features.

Bad:

This pull request implements functionality for YAML frontmatter parsing in the issue template processing subsystem. The implementation enhances the system's capabilities regarding template format support.

Write like a human talking to another human, not like a technical specification.

Error Handling

When operations fail, provide clear context:

  • What was being attempted
  • What went wrong
  • What the user should do to fix it

Example:

Failed to create PR: current branch has no commits ahead of main.

Make sure you've committed your changes and that your branch differs from the base branch.

Confirmation and Feedback

For destructive or significant operations:

  • Show what will happen before executing
  • Ask for confirmation when appropriate
  • Provide clear feedback after completion

For search operations with --check:

  • Display found items clearly
  • Let the user decide whether to proceed
  • Don't automatically skip creation if duplicates exist

Additional Resources

For detailed workflows, examples, and edge case handling, refer to these reference documents:

  • ./references/pull-requests.md - Complete pull request workflow including git state validation, semantic analysis strategies, template examples, and reviewer management
  • ./references/issues.md - Complete issue creation workflow including template parsing, label application strategies, and duplicate detection
  • ./references/issues-claude-code.md - Claude Code-specific issue creation including all template formats, environment gathering scripts, and submission guidelines
  • ./references/issues-codex-cli.md - Codex CLI-specific issue creation including bug reports, feature requests, docs issues, and VS Code extension templates
  • ./references/issues-biome.md - Biome-specific issue creation including playground reproduction links, formatter/linter bug templates, and environment detection
  • ./references/discussions.md - GitHub discussions workflow including GraphQL queries, category selection logic, and search strategies

These references provide implementation details, code examples, and troubleshooting guidance for each workflow type.

平台分布

trae

60.48%
按下载量换算44

Claude Code

29.67%
按下载量换算21

安全审计

暂无安全审计结果可展示。

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add paulrberg/dot-claude --skill "oss" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills