Token导航 LogoToken导航

Agent Skill · 研究检索

jira(Jira 项目管理)

用于处理 Jira 项目、任务、缺陷、Sprint、负责人和状态流转。它适合让 Agent 辅助查询工单、汇总迭代进展、创建任务或整理需求和缺陷信息。使用时要确认项目权限、字段配置和工作流规则,不同团队的 Issue 类型、状态和必填字段可能不同;涉及批量改状态、改负责人或创建工单时,应先预览变更内容再执行。

jira 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,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 34,372。

总安装 2,233周安装 94GitHub Stars 公开资料未说明下载 782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:jira(Jira 项目管理)
来源仓库:https://github.com/jorgemuza/orbit
仓库路径:skills/jira
安装命令:
npx skills add https://github.com/jorgemuza/orbit --skill jira
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/jorgemuza/orbit --skill jira

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。

jira 用于处理 Jira 项目、任务、缺陷和 Sprint 管理。适用于查询工单、汇总迭代进展和创建任务。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。通过 Jira API 提供工单管理和状态流转支持。

使用时需确认字段配置和工作流规则,预览变更后再执行批量操作。jira 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/jira

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/jorgemuza/orbit --skill jira;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

jira 怎么安装?

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

jira 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。

jira 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。

jira 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。

jira 用的人多吗?

公开数据是累计安装 2,233 次,近一周安装 94 次,下载 782 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。

jira 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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
jira

Jira 项目管理 · 本页

2,233—接触敏感数据未明确未通过

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

Jira with orbit CLI

Manage Jira issues, epics, sprints, boards, projects, releases, custom fields, screens, statuses, and issue types using the orbit CLI. Supports both Jira Cloud (REST API v3) and Jira Server/Data Center (REST API v2) with multi-profile support and 1Password secret resolution.

Prerequisites

  1. orbit CLI installed — if which orbit fails, install with:

- macOS/Linux (Homebrew): brew install jorgemuza/tap/orbit - macOS/Linux (script): curl -sSfL https://raw.githubusercontent.com/jorgemuza/orbit/main/install.sh | sh - Windows (Scoop): scoop bucket add jorgemuza https://github.com/jorgemuza/scoop-bucket && scoop install orbit

  1. A profile with a jira-cloud or jira-onprem service configured in ~/.config/orbit/config.yaml
  2. Valid credentials — API token for Cloud (basic auth with email + token), PAT for Server
  3. Credentials can be stored in 1Password with op:// prefix for automatic resolution

Quick Reference

All commands follow the pattern: orbit -p <profile> jira <resource> <action> [flags]

The -o flag controls output format: table (default), json, yaml.

For full command details and flags, see references/commands.md. For Jira wiki markup formatting (Server only), see references/wiki-markup.md.

Core Workflows

Viewing Issues

# View issue with comments (shows last 10 by default)
orbit -p myprofile jira issue view PROJ-123

# View with more comments
orbit -p myprofile jira issue view PROJ-123 --comments 50

# JSON output (includes full data + all comments)
orbit -p myprofile jira issue view PROJ-123 -o json

The view command shows: summary, type, status, priority, assignee, reporter, labels, components, description, subtasks, links, attachments (with filename, size, download URL), and comments with author/date/edit timestamps. Attachments appear before comments so screenshots referenced in comment text have context.

Creating Issues

# Simple story
orbit -p myprofile jira issue create --project PROJ --type Story --summary "Add login page"

# Bug with priority and assignment
orbit -p myprofile jira issue create --project PROJ --type Bug \
  --summary "Fix timeout" --priority High --assignee john.doe

# Sub-task under a parent
orbit -p myprofile jira issue create --project PROJ --type Sub-task \
  --parent PROJ-123 --summary "Add validation"

Creating Epics

Epics require the "Epic Name" custom field. The orbit CLI auto-sets this from the summary when --type Epic is used. Use --epic-name to override.

# Create an epic (Epic Name auto-set from summary)
orbit -p myprofile jira issue create --type Epic --project PROJ \
  --summary "Q1 Auth Revamp" --priority Highest

# Create epic with Parent Link to a Capability/Initiative
orbit -p myprofile jira issue create --type Epic --project PROJ \
  --summary "Okta Authentication Foundation" \
  --field "customfield_27521=PRT-4378" \
  --priority Highest

The --field flag sets arbitrary custom fields as key=value. Use jira field list to discover field IDs.

Editing Issues with Formatted Descriptions

Jira Server uses wiki markup, not Markdown. Jira Cloud uses ADF (handled automatically by orbit for plain text). Always format --body values using wiki markup syntax for Server instances.

orbit -p myprofile jira issue edit PRT-123 --body "h2. Value Statement

The platform provides real authentication via Okta.

h2. User Stories

* *Story 1:* As a platform admin, I want all API requests to require a valid Okta JWT.
** Okta custom AS configured with epsilon_claims claim
** AUTH_BYPASS removed from all Terraform task definitions"

Editing Custom Field Values

The -F flag supports multiple value formats. The correct format depends on the Jira field type:

# Single-select field — use value: shorthand
orbit -p myprofile jira issue edit PROJ-123 -F customfield_10397=value:Yes

# Multi-select field — MUST use JSON array of objects (value: shorthand will return 400)
orbit -p myprofile jira issue edit PROJ-123 -F 'customfield_10398=[{"value":"Developer"},{"value":"QA"}]'

# Multi-select with a single value still needs array format
orbit -p myprofile jira issue edit PROJ-123 -F 'customfield_10400=[{"value":"Confirmed"}]'

# Number field
orbit -p myprofile jira issue edit PROJ-123 -F customfield_10399=3

# Plain string field
orbit -p myprofile jira issue edit PROJ-123 -F customfield_10010=my-value

Important: Multi-select fields (checkboxes, multi-select dropdowns) require [{"value":"X"}] array format. Using value:X on a multi-select field returns a 400 error.

Transitioning Issues

# Move to In Progress
orbit -p myprofile jira issue move PROJ-123 "In Progress"

# Close with comment and resolution
orbit -p myprofile jira issue move PROJ-123 Done --comment "Fixed in v2.1" --resolution Fixed

Comments with @mentions

Use [~accountId:ID] to @mention users in comments (Cloud). Jira renders these as clickable user links. For Server, use [~username].

# Add a comment mentioning the assignee
orbit -p myprofile jira issue comment PROJ-123 -b "Hey [~accountId:5b10ac8d82e05b22cc7d4ef5], this needs rework"

# Look up a user's accountId
orbit -p myprofile jira user search "Jorge"

Attachments

# Attach a file to an issue
orbit -p myprofile jira issue attach PROJ-123 report.pdf

# Attach multiple files
orbit -p myprofile jira issue attach PROJ-123 screenshot.png log.txt

# List attachments
orbit -p myprofile jira issue attachments PROJ-123

# Download all attachments
orbit -p myprofile jira issue download PROJ-123

# Download only screenshots/images
orbit -p myprofile jira issue download PROJ-123 --images-only --output ./screenshots

# Remove an attachment by ID
orbit -p myprofile jira issue detach 12345

Searching Issues

# Filter by project and type
orbit -p myprofile jira issue list --project PROJ --type Epic

# Filter by assignee and status
orbit -p myprofile jira issue list --assignee me --status "In Progress"

# Raw JQL query
orbit -p myprofile jira issue list --jql "project = PROJ AND sprint in openSprints()"

# JSON output for processing
orbit -p myprofile jira issue list --project PROJ -o json

# Fresh list — bypasses the JQL search index (useful right after transitions)
orbit -p myprofile jira issue list --jql "project = PROJ AND status = Open" --fresh
JQL index lag — issue list hits Jira's JQL search index, which is eventually consistent and can trail transitions by 30–60s. When the user asks for a "current", "fresh", or "live" list, or they just transitioned a ticket in the same session, pass --fresh so each row is re-verified against the authoritative single-issue endpoint. Without --fresh, list results are informational; always re-verify a single key with jira issue view KEY before acting on it.

Exporting Epic Hierarchies

Export full epic hierarchies (epic → stories → sub-tasks) to structured markdown directories.

# Export a single epic
orbit -p myprofile jira export BS-1243 -O ./jira-export

# Export multiple epics
orbit -p myprofile jira export BS-1243 BS-1255 BS-1246 -O ./jira-export

Output structure:

<outdir>/<EPIC-KEY>/INDEX.md                    — epic metadata + story list
<outdir>/<EPIC-KEY>/<STORY-KEY>/INDEX.md        — story metadata + sub-task list
<outdir>/<EPIC-KEY>/<STORY-KEY>/<TASK-KEY>.md   — individual sub-task file

Each file includes a metadata table (type, status, assignee, priority, labels, story points) and the full description converted from Jira ADF to markdown. Use --max-results to control how many children are fetched per level (default: 100).

Sprint Management

# List sprints on a board
orbit -p myprofile jira sprint list --board-id 42
orbit -p myprofile jira sprint list --board-id 42 --state active

# View sprint details
orbit -p myprofile jira sprint view 1366

# List issues in a sprint
orbit -p myprofile jira sprint list 1366

# Create a sprint
orbit -p myprofile jira sprint create --board-id 42 --name "Sprint 5" --start-date 2026-04-01 --end-date 2026-04-15

# Start a sprint
orbit -p myprofile jira sprint start 1366

# Close a sprint
orbit -p myprofile jira sprint close 1366

# Update sprint name/dates/goal
orbit -p myprofile jira sprint update 1366 --name "Sprint 5 Extended" --end-date 2026-04-20

# Add issues to a sprint
orbit -p myprofile jira sprint add 1366 PROJ-101 PROJ-102

# Move issues to backlog
orbit -p myprofile jira sprint remove PROJ-101 PROJ-102

# Delete a sprint
orbit -p myprofile jira sprint delete 1366

Dashboards & Gadgets

Create and manage Jira dashboards with gadgets programmatically.

# List dashboards
orbit -p myprofile jira dashboard list
orbit -p myprofile jira dashboard list --filter "Sprint"

# Create a dashboard
orbit -p myprofile jira dashboard create --name "AI Metrics" --description "AI adoption tracking"

# View dashboard details
orbit -p myprofile jira dashboard view 10408

# Delete a dashboard
orbit -p myprofile jira dashboard delete 10408

# Add gadgets to a dashboard (use URI format for Cloud)
orbit -p myprofile jira dashboard gadget add 10408 \
  --uri "rest/gadgets/1.0/g/com.atlassian.jira.gadgets:filter-results-gadget/gadgets/filter-results-gadget.xml" \
  --title "My Filter Results" --color blue

# List gadgets on a dashboard
orbit -p myprofile jira dashboard gadget list 10408

# Remove a gadget
orbit -p myprofile jira dashboard gadget remove 10408 35501

# Configure gadget properties (e.g., bind a filter)
orbit -p myprofile jira dashboard gadget property set 10408 35501 config '{"filterId":"11233"}'
orbit -p myprofile jira dashboard gadget property get 10408 35501 config
orbit -p myprofile jira dashboard gadget property list 10408 35501

Filters

Create, search, and manage saved JQL filters.

# List favourite filters
orbit -p myprofile jira filter list

# Search all accessible filters
orbit -p myprofile jira filter search
orbit -p myprofile jira filter search --name "Sprint"

# View filter details
orbit -p myprofile jira filter view 10195

# Create a filter
orbit -p myprofile jira filter create --name "My Bugs" --jql "project = PROJ AND type = Bug" --favourite

# Update a filter
orbit -p myprofile jira filter update 10195 --jql "project = PROJ AND sprint in openSprints()"
orbit -p myprofile jira filter update 10195 --name "Active Sprint"

# Delete a filter
orbit -p myprofile jira filter delete 12345

Field Management (Cloud Only)

Manage custom fields, their contexts, and options programmatically.

Listing Fields

# List all fields
orbit -p myprofile jira field list

# List only custom fields
orbit -p myprofile jira field list --custom

# Filter by name or ID
orbit -p myprofile jira field list --filter "AI"

Creating Custom Fields

# Create a select field (shorthand types: select, multiselect, number, checkbox, text, textarea)
orbit -p myprofile jira field create --name "AI Assisted" --type select \
  --description "Was AI used on this ticket?"

# Create a number field
orbit -p myprofile jira field create --name "AI Prompt Iterations" --type number \
  --description "How many prompt cycles to get working output"

# Create a checkbox field
orbit -p myprofile jira field create --name "Human Review Confirmed" --type checkbox \
  --description "Engineer confirms AI output was reviewed"

Managing Field Contexts and Options

# List field contexts
orbit -p myprofile jira field context-list customfield_10397

# List existing options
orbit -p myprofile jira field option-list customfield_10397 10817

# Add options to a select/multiselect field
orbit -p myprofile jira field option-add customfield_10397 10817 \
  --values "Yes,No,Partial"

Screen Management

Control which fields appear on issue create/edit/view forms by managing screen tabs and field assignments.

Listing Screens and Tabs

# List all screens
orbit -p myprofile jira screen list

# Filter screens by name
orbit -p myprofile jira screen list --filter "PYMT"

# List tabs on a screen
orbit -p myprofile jira screen tab-list 10089

# List fields on a screen tab
orbit -p myprofile jira screen field-list 10089 10189

Creating Tabs

# Create a new tab to group related fields
orbit -p myprofile jira screen tab-create 10089 "AI Workflow"

Adding and Removing Fields

# Add fields to a screen tab
orbit -p myprofile jira screen field-add 10089 10868 \
  --fields "customfield_10397,customfield_10398,customfield_10399"

# Remove fields from a screen tab
orbit -p myprofile jira screen field-remove 10089 10189 \
  --fields "customfield_10397,customfield_10398"

Moving Fields Between Tabs

# Move fields from General tab to AI Workflow tab
orbit -p myprofile jira screen field-move 10089 10189 10868 \
  --fields "customfield_10397,customfield_10398,customfield_10399"

Status and Issue Type Management

Listing Workflow Statuses

# List all statuses with their categories
orbit -p myprofile jira status list

# Filter with grep
orbit -p myprofile jira status list | grep -i "review"

Listing Issue Types

orbit -p myprofile jira issuetype-list

Deleting Issues

Delete issues with issue delete. Use --cascade to also delete all subtasks.

Important: There is no --confirm or --force flag. The command deletes immediately.

# Delete a single issue
orbit -p myprofile jira issue delete PROJ-123

# Delete an issue and all its subtasks
orbit -p myprofile jira issue delete PROJ-100 --cascade

Deleting a hierarchy (parent → children → grandchildren): Delete bottom-up to avoid orphan errors, or use --cascade at each level that has subtasks.

# Example: delete Story (with sub-tasks) → Epic → Capability
orbit -p myprofile jira issue delete PROJ-50 --cascade   # Story + sub-tasks
orbit -p myprofile jira issue delete PROJ-40              # Epic (now childless)
orbit -p myprofile jira issue delete PRT-200              # Capability (now childless)

Tip: --cascade only deletes Jira sub-tasks (child issues of type Sub-task). It does NOT delete Stories under an Epic or Epics under a Capability. For full hierarchy deletion, query children with JQL first, then delete bottom-up.

# Find all children of an issue
orbit -p myprofile jira issue list --jql '"Parent" = PROJ-40' -o json

Important Notes

  • Cloud vs Server — Use service type jira-cloud for Atlassian Cloud (uses API v3 with ADF for descriptions). Use jira-onprem for Server/Data Center (uses API v2 with wiki markup).
  • Description formatting — For Server instances, always use Jira wiki markup for --body values. For Cloud, orbit auto-converts plain text to ADF. See references/wiki-markup.md for wiki markup syntax.
  • Field management — field create, field context-list, field option-list, and field option-add are Cloud-only features.
  • Screen management — Adding fields to screens makes them appear on issue create/edit forms. Use tab-create to group related fields into a dedicated tab.
  • Epic type cannot use --parent flag — Jira rejects it because Epic is not a sub-task type. Use the --field "customfield_27521=KEY" (Parent Link) instead.
  • 1Password integration — Credentials in config can use op://vault/item/field and are resolved at runtime. Run orbit auth once to resolve and cache all secrets for 8 hours (single biometric prompt). Use orbit auth clear to wipe the cache. Without orbit auth, secrets are still resolved on each command but may trigger repeated biometric prompts.

平台分布

Codex

40.08%
按下载量换算313

Claude

28.36%
按下载量换算222

Cursor

19.25%
按下载量换算151

Gemini CLI

9.07%
按下载量换算71

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

未通过

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