jira-epic-to-stories

JIRA Epic to Stories Skill

This skill teaches Copilot how to automatically create JIRA stories from epic descriptions using the jira CLI.

Prerequisites

This skill requires the Jira CLI (ankitpokhrel/jira-cli) to be installed and properly authenticated on the host machine. The scripts within this skill invoke the jira binary directly.

When to Use This Skill

Activate this skill when:

• User asks to "create stories from epic"
• User provides an epic key and wants to generate child stories
• User mentions "parse epic description" or "generate stories"

Epic Format Expected

The skill parses epics with this description structure:

**Stories**:

• First story title
    • (optional) Additional context or sub-task
• Second story title
    • (optional) Additional context

Also supports:

• Stories: (without bold)
• Different indentation (2-6 spaces for bullets)
• ANSI escape codes in CLI output

Besides Stories, there are other issues that can be created from epics, such as:

• Jobs: These are identified as Steps or Progression

These are also to be created as jira issues using the type Job

Implementation Pattern

Step 1: Fetch Epic Content

import subprocess
import time

process = subprocess.Popen(
    ["jira", "issue", "view", epic_key],
    stdin=subprocess.PIPE,
    stdout=subprocess.PIPE,
    stderr=subprocess.PIPE,
    text=True
)

# Wait for content to load
time.sleep(3)

# Send 'q' to quit pager
stdout, stderr = process.communicate(input='q', timeout=5)

Step 2: Parse Stories from Description

import re

def parse_stories(epic_content: str) -> list:
    """Extract story items from epic description"""
    stories = []

    # Remove ANSI escape codes
    clean_content = re.sub(r'\x1b\[[0-9;]+m', '', epic_content)
    lines = clean_content.split('\n')

    in_stories_section = False
    current_story = None

    for line in lines:
        # Find Stories section (with or without bold)
        if re.search(r'\*?\*?Stories\*?\*?:', line):
            in_stories_section = True
            continue

        # Exit at next section
        if in_stories_section and ('————' in line or 'Linked Issues' in line):
            break

        if in_stories_section:
            # Top-level bullet (2 spaces): "  • Story title"
            match = re.match(r'^(\s{2})•\s+(.+)', line)
            if match:
                if current_story:
                    stories.append(current_story)

                text = match.group(2).strip()
                current_story = {
                    'summary': text,
                    'description': text
                }
            # Sub-bullet (6 spaces): "      • Additional context"
            elif current_story:
                sub_match = re.match(r'^\s{6,}•\s+(.+)', line)
                if sub_match:
                    context = sub_match.group(1).strip()
                    current_story['description'] += f"\n\n{context}"

    if current_story:
        stories.append(current_story)

    return stories

Step 3: Create Stories

def create_story(epic_key: str, summary: str, description: str) -> tuple:
    """Create a story as child of epic"""

    process = subprocess.Popen(
        ["jira", "issue", "create",
         "-t", "Story",
         "-P", epic_key,
         "-s", summary,
         "-b", description],
        stdin=subprocess.PIPE,
        stdout=subprocess.PIPE,
        stderr=subprocess.PIPE,
        text=True
    )

    # Send enter to submit (select default "Submit" option)
    stdout, stderr = process.communicate(input="\n", timeout=30)

    if process.returncode == 0:
        # Extract URL from output
        match = re.search(r'https://[^\s]+', stdout)
        if match:
            return True, match.group(0)

    return False, stderr

Complete Automation Script

A full Python script is available in /jira/epic-to-stories.py with:

# Create stories from epic
python3 jira/epic-to-stories.py DEVOPS-43103

# Preview without creating (dry-run)
python3 jira/epic-to-stories.py DEVOPS-43103 --dry-run

# Debug mode to see raw content
python3 jira/epic-to-stories.py DEVOPS-43103 --debug

Key Implementation Details

Handling ANSI Escape Codes

JIRA CLI output includes formatting codes. Always clean them:

clean_text = re.sub(r'\x1b\[[0-9;]+m', '', raw_output)

Handling Interactive Prompts

The CLI shows an interactive menu after jira issue create. Handle it:

# Automatically submit by sending enter
stdout, stderr = process.communicate(input="\n", timeout=30)

Parsing Indentation

Stories use 2-space indent, sub-bullets use 6+ spaces:

# Main story: "  • Title"
main_match = re.match(r'^(\s{2})•\s+(.+)', line)

# Sub-bullet: "      • Context"
sub_match = re.match(r'^\s{6,}•\s+(.+)', line)

Error Handling

Always handle:

• Timeout errors when fetching epics
• Failed story creation (report which ones failed)
• Invalid epic keys
• Empty story lists

Output Format

Provide clear output:

============================================================
Found 2 stories to create:
============================================================

1. Create ACP SRE Agent Plugin with Incident Investigation Capabilities
2. Adds Incident Remediation capabilities to agent for quick incident recovery

============================================================

✓ Created: https://jira.alauda.cn/browse/DEVOPS-43119
✓ Created: https://jira.alauda.cn/browse/DEVOPS-43120

============================================================
Summary:
============================================================
Created: 2/2 stories

Dry-Run Mode

Always support --dry-run flag to preview without creating:

[DRY RUN] Would create story:
  Summary: Story title
  Description: Story description with context

Validation Steps

Before creating stories:

1. Validate epic key format (PROJECT-NUMBER)
2. Verify epic exists and is accessible
3. Check if stories already exist as children
4. Confirm at least one story found in description

Resources

• Python script: /jira/epic-to-stories.py
• Shell script: /jira/epic-to-stories-skill.sh
• Documentation: /jira/epic-to-stories.md