Token导航 LogoToken导航

Agent Skill · 研究检索

atlassian(atlassian 命令行)

atlassian 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。

atlassian 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,权限为接触敏感数据;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 64,068。

总安装 306周安装 13GitHub Stars 公开资料未说明下载 107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
unknown

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:atlassian(atlassian 命令行)
来源仓库:https://github.com/boazy/skills
仓库路径:skills/atlassian
安装命令:
npx skills add https://github.com/boazy/skills --skill atlassian
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。

skills.shnpx skills
npx skills add https://github.com/boazy/skills --skill atlassian

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。

atlassian 提供 Jira 和 Confluence 的 REST API 访问能力。支持 Issue 和页面的完整 CRUD 操作,无需额外 MCP 服务器。需配置 ATLASSIAN_SITE、ATLASSIAN_EMAIL 和 ATLASSIAN_API_TOKEN 环境变量。

涉及生产系统读写,应确保 API token 权限最小化并避免泄露。atlassian 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/atlassian

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 https://github.com/boazy/skills --skill atlassian;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

    它需要接触敏感数据,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

atlassian 怎么安装?

在终端执行 npx skills add https://github.com/boazy/skills --skill atlassian。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

atlassian 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。

atlassian 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

atlassian 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

atlassian 用的人多吗?

公开数据是累计安装 306 次,近一周安装 13 次,下载 107 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。

atlassian 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 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
atlassian

atlassian 命令行 · 本页

306—接触敏感数据未明确有提醒

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

Atlassian Skill

Access Jira and Confluence directly via REST APIs. This skill provides full CRUD operations for issues and pages without requiring the MCP server.

Authentication

Requires environment variables in ~/.local/secrets/atlassian.env:

Available Scripts

Jira

Search Issues
bunx tsx scripts/jira-search.ts "<JQL query>" [maxResults] [nextPageToken]

Examples:

  • bunx tsx scripts/jira-search.ts "assignee = currentUser() AND status!= Done"
  • bunx tsx scripts/jira-search.ts "project = PROJ AND type = Bug" 50
  • bunx tsx scripts/jira-search.ts "project = PROJ" 50 "token..." (pagination)

See docs/jql-guide.md for JQL syntax reference.

Get Issue Details
bunx tsx scripts/jira-get.ts <issueKey>

Example: bunx tsx scripts/jira-get.ts PROJ-123

Create Issue
bunx tsx scripts/jira-create.ts '<JSON>'

Single issue:

bunx tsx scripts/jira-create.ts '{"project": "PROJ", "type": "Story", "summary": "New feature", "description": "Details here"}'

Bulk create (array):

bunx tsx scripts/jira-create.ts '[{"project": "PROJ", "type": "Bug", "summary": "Bug 1"}, {"project": "PROJ", "type": "Bug", "summary": "Bug 2"}]'
Update Issue
bunx tsx scripts/jira-update.ts <issueKey> '<JSON updates>'

Example: bunx tsx scripts/jira-update.ts PROJ-123 '{"status": "In Progress", "assignee": "user@example.com"}'

Comments
# Get comments
bunx tsx scripts/jira-comment.ts <issueKey> get

# Add comment
bunx tsx scripts/jira-comment.ts <issueKey> add "<comment text>"
Upload Attachment
bunx tsx scripts/jira-attachment.ts <issueKey> <filePath> [fileName]

Examples:

  • bunx tsx scripts/jira-attachment.ts PROJ-123./diagram.png
  • bunx tsx scripts/jira-attachment.ts PROJ-123./image.png architecture.png

Image embedding note:

  • For markdown image syntax (![alt](url)), the converter uses mediaSingle only for Atlassian-hosted URLs.
  • External image URLs fall back to a clickable link to avoid Jira INVALID_INPUT errors.
  • If you want embedded images, upload the file first with jira-attachment.ts, then use the returned Atlassian contentUrl in your markdown image URL.

Confluence

Personal Space

When the user asks to work with their personal space (e.g., "write a page in my personal space", "search my personal space", "read a page from my space"), you MUST first discover their personal space before performing the requested operation.

Auto-detection rule: Any mention of "my space", "my personal space", "personal space", or "my Confluence space" means the user's personal Confluence space. Always run the discovery script first to get the space key and ID, then use those values in subsequent operations (create, search, get, etc.).

Discover Personal Space
bunx tsx scripts/confluence-personal-space.ts

No arguments needed. Returns the current user's personal space key, ID, name, and URL.

Example output:

{
  "accountId": "5b10a2844c20165700ede21g",
  "displayName": "Jane Smith",
  "space": {
    "id": 98304,
    "key": "~5b10a2844c20165700ede21g",
    "name": "Jane Smith",
    "type": "personal",
    "status": "current",
    "url": "https://yourcompany.atlassian.net/wiki/spaces/~5b10a2844c20165700ede21g"
  }
}

Then use the returned space.key as the space key for other Confluence operations. For example:

  • Create page: bunx tsx scripts/confluence-create.ts '{"space": "~5b10a2844c20165700ede21g", "title": "My Notes", "body": "<p>Content</p>"}'
  • Search pages: bunx tsx scripts/confluence-search.ts "space = ~5b10a2844c20165700ede21g AND type = page"
  • Get page by title: bunx tsx scripts/confluence-get.ts "My Notes" ~5b10a2844c20165700ede21g

Note: Personal space keys on Confluence Cloud use the format ~accountId (not ~username). The discovery script handles this automatically.

Search Pages
bunx tsx scripts/confluence-search.ts "<CQL query>" [maxResults]

Examples:

  • bunx tsx scripts/confluence-search.ts "title ~ 'Roadmap'"
  • bunx tsx scripts/confluence-search.ts "space = DEV AND type = page" 25

See docs/cql-guide.md for CQL syntax reference.

Get Page Content
bunx tsx scripts/confluence-get.ts <pageId>
# or by title
bunx tsx scripts/confluence-get.ts "<page title>" <spaceKey>
Create Page
bunx tsx scripts/confluence-create.ts '<JSON>'

Example:

bunx tsx scripts/confluence-create.ts '{"space": "DEV", "title": "New Page", "body": "<p>Content here</p>"}'

Optional parent page:

bunx tsx scripts/confluence-create.ts '{"space": "DEV", "title": "Child Page", "body": "<p>Content</p>", "parentId": "123456"}'
Update Page
bunx tsx scripts/confluence-update.ts <pageId> '<JSON updates>'

Example: bunx tsx scripts/confluence-update.ts 123456 '{"title": "Updated Title", "body": "<p>New content</p>"}'

Page Properties (v2)
# List properties
bunx tsx scripts/confluence-properties.ts <pageId> get

# Get a single property
bunx tsx scripts/confluence-properties.ts <pageId> get <propertyKey>

# Set a property
bunx tsx scripts/confluence-properties.ts <pageId> set '<JSON>'

# Delete a property
bunx tsx scripts/confluence-properties.ts <pageId> delete <propertyKey>

# Get page emoji
bunx tsx scripts/confluence-properties.ts <pageId> get-emoji

# Set page emoji (updates both emoji-title-published and emoji-title-draft)
bunx tsx scripts/confluence-properties.ts <pageId> set-emoji "🚀"

# Remove page emoji
bunx tsx scripts/confluence-properties.ts <pageId> remove-emoji

Example property payloads:

  • '{"key": "my-property", "value": {"foo": "bar"}}'
  • '{"properties": [{"key": "one", "value": 1}, {"key": "two", "value": "two"}]}'

Note: Page CRUD uses the Confluence REST API v2. CQL search still uses the legacy endpoint because the v2 API does not expose CQL search.

Query Language References

For generating correct queries:

  • Jira: Read docs/jql-guide.md for JQL syntax, fields, operators, and functions
  • Confluence: Read docs/cql-guide.md for CQL syntax and fields

Large Content — File-Based Input

All scripts that accept a '<JSON>' argument also support @<filepath> syntax: write the JSON to a file first, then pass @path/to/file.json instead of the inline JSON string. This avoids shell argument-length limits that cause failures with large page bodies or issue descriptions.

You MUST use file-based input when creating or updating Confluence pages or Jira issues with non-trivial body/description content. Inline JSON is fine only for short payloads (simple metadata updates, status changes, etc.).

Workflow

  1. Write the JSON payload to a temporary file (e.g., /tmp/confluence-payload.json)
  2. Pass @/tmp/confluence-payload.json as the argument instead of the raw JSON string
  3. Clean up the temp file afterward

Examples

Create a Confluence page with a large body:

# 1. Write payload to file
cat > /tmp/page.json << 'ENDJSON'
{"space": "DEV", "title": "Architecture Overview", "body": "<h1>Architecture</h1><p>Long content here...</p>"}
ENDJSON

# 2. Pass with @filepath
bunx tsx scripts/confluence-create.ts @/tmp/page.json

Update a Confluence page:

bunx tsx scripts/confluence-update.ts 123456 @/tmp/update-payload.json

Create a Jira issue with a long description:

bunx tsx scripts/jira-create.ts @/tmp/issue.json

Update a Jira issue:

bunx tsx scripts/jira-update.ts PROJ-123 @/tmp/update.json

Important: When generating content programmatically (as an AI agent), always use the Write tool to create the JSON file, then invoke the script with @filepath. Never attempt to pass large HTML or markdown content as an inline shell argument.

Common Workflows

Find and update my open issues

  1. Search: bunx tsx scripts/jira-search.ts "assignee = currentUser() AND status!= Done"
  2. Update: bunx tsx scripts/jira-update.ts PROJ-123 '{"status": "Done"}'

Create issues from a list

  1. Bulk create: bunx tsx scripts/jira-create.ts '[{...}, {...}, {...}]'

Find and read documentation

  1. Search: bunx tsx scripts/confluence-search.ts "title ~ 'API Documentation'"
  2. Get content: bunx tsx scripts/confluence-get.ts 123456

Create a new documentation page

  1. Create: bunx tsx scripts/confluence-create.ts '{"space": "DEV", "title": "API Guide", "body": "<h1>API Guide</h1><p>...</p>"}'

Work with your personal space

  1. Discover: bunx tsx scripts/confluence-personal-space.ts → note the space.key value
  2. Search: bunx tsx scripts/confluence-search.ts "space = <space.key> AND type = page"
  3. Create: bunx tsx scripts/confluence-create.ts '{"space": "<space.key>", "title": "My Page", "body": "<p>Content</p>"}'
  4. Read: bunx tsx scripts/confluence-get.ts "<page title>" <space.key>

平台分布

Codex

36.96%
按下载量换算40

Claude

29.29%
按下载量换算31

Cursor

21.67%
按下载量换算23

Gemini CLI

10.07%
按下载量换算11

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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