AgentSkillsCN

jira-ticket-reader

阅读Jira工单，并根据信息与起始提示创建任务文件夹。指令：/read-ticket、/task-info、/import-jira。触发词：“阅读工单”、“获取工单信息”、“记录工单”、“获取Jira问题”、“导入工单”、“加载工单详情”。

SKILL.md
--- frontmatter
name: jira-ticket-reader
description: Read Jira ticket and create task folder with info + starting prompt. Commands: /read-ticket, /task-info, /import-jira. Triggers: "read ticket", "get ticket info", "document ticket", "fetch jira issue", "import ticket", "load ticket details".
license: MIT
metadata:
  author: metrc-tech
  version: "1.5.0"

Jira Ticket Reader

Extracts Jira ticket information from the current git branch, fetches complete ticket data, and creates a task folder with ticket documentation and a starting prompt for LLM-assisted development.

How It Works

  1. •Validate prerequisites - Verify git repository and Jira connection
  2. •Extract ticket ID - From branch name (pattern: [A-Z]+-\d+) or query Jira
  3. •Fetch ticket data - Retrieve complete issue details with single API call
  4. •Convert to Markdown - Transform rich text format to readable documentation
  5. •Generate files - Create ticket documentation and starting prompt using templates
  6. •Validate output - Verify files created successfully
  7. •Report success - Confirm with file locations and links

Usage

Prerequisites

  • •Git repository with branch containing ticket ID
  • •Jira API access with read permissions
  • •Write access to local documentation directory

Workflow

1. Validate Environment

  • •Check git is available and repository exists
  • •Verify Jira API connection is active
  • •Confirm current directory has write permissions

See Git Operations for validation commands.

2. Get Jira Connection

  • •Retrieve available Jira instances
  • •Select appropriate cloud/server instance
  • •Extract connection identifier

See Jira MCP API Reference for connection details.

3. Determine Ticket ID

Option A: From Branch Name

  • •Get current branch name
  • •Extract ticket ID using pattern: [A-Z]+-\d+
  • •Example: fix/MCA-402-bug-fix → MCA-402

Option B: Query Jira

  • •If no ticket in branch, query for assigned tickets in "In Development" status
  • •Present list to user if multiple tickets found
  • •Accept manual input if no tickets found

See Git Operations for branch extraction.

4. Fetch Complete Ticket Data

  • •Request all relevant fields in single API call
  • •Include: summary, description, status, priority, type, labels, components, reporter, assignee, timestamps, comments, attachments
  • •Use field expansion for complete data

See Jira MCP API Reference for API details.

5. Convert Rich Text to Markdown

  • •Parse structured document format (ADF)
  • •Apply conversion rules for all element types
  • •Handle nested structures and formatting marks
  • •Gracefully degrade unsupported elements

See ADF Conversion Guide for conversion details and company-specific fields.

6. Generate Documentation Files

Create directory structure: personal/tasks/{TICKET-ID}/

File 1: {TICKET-ID}.md - Ticket information

  • •Streamlined ticket info with only relevant details
  • •Issue Details, Steps to Reproduce, Technical Context
  • •Comments (most recent first)
  • •Attachments list

See Ticket Info Template for structure.

File 2: {TICKET-ID}.prompt.md - Starting prompt

  • •References the {TICKET-ID}.md file for full details
  • •Problem Statement, Objective, Constraints
  • •Request for AI assistance with structured approach

See Prompt Template for structure.

File 3: images/ - Downloaded attachments (CONDITIONAL)

  • •Only create if attachments were successfully downloaded
  • •Do NOT create folder if no attachments or download fails
  • •Reference downloaded images in {TICKET-ID}.md

See Jira MCP Reference for attachment download details.

7. Validate Files

  • •Verify all files created
  • •Check file sizes are non-zero
  • •Confirm valid text encoding

8. Report Results

code
🔍 Ticket ID: {TICKET-ID}
✅ Ticket Info: personal/tasks/{TICKET-ID}/{TICKET-ID}.md
🚀 Starting Prompt: personal/tasks/{TICKET-ID}/{TICKET-ID}.prompt.md
�️ Images: personal/tasks/{TICKET-ID}/images/ (if attachments downloaded)
🔗 Jira: https://[instance]/browse/{TICKET-ID}

Output

Creates task folder in personal/tasks/{TICKET-ID}/:

  • •{TICKET-ID}.md - Streamlined ticket information (Issue Details, Steps to Reproduce, Technical Context, Comments, Attachments)
  • •{TICKET-ID}.prompt.md - Starting prompt that references the ticket file, ready to begin development
  • •images/ - Downloaded attachments (only if successfully retrieved)

Error Handling

Retry Strategy:

  • •Retry failed API calls with exponential backoff
  • •Maximum 3 attempts per operation
  • •Backoff intervals: 1s, 2s, 4s
  • •Timeout: 30 seconds per attempt

Common Issues:

IssueResolution
No ticket in branchQuery Jira or prompt user for ticket ID
Connection failedVerify API credentials and retry
Ticket not foundConfirm ticket exists and user has access
Empty contentCheck alternate fields; include comments
File creation failedVerify directory permissions

See Troubleshooting Guide for detailed solutions.

Document Format Conversion

Basic element mapping:

Source ElementTarget Format
ParagraphPlain text with line breaks
Heading (1-6)Markdown headings # to ######
Bullet listMarkdown list with -
Numbered listMarkdown list with 1.
Code blockFenced code blocks with language
Bold text**text**
Italic text*text*
Inline code`text`

Fallback Handling:

  • •Log warnings for unsupported elements
  • •Extract plain text when possible
  • •Add comments for skipped content
  • •Continue processing (never fail completely)

See ADF Conversion Guide for complete conversion rules and company-specific examples.

References

Load these resources on-demand for detailed implementation: