jira-parser

name	jira-parser
description	Parse and normalize Jira issue fields into a structured format for STP generation
model	claude-opus-4-6

Jira Parser Skill

Phase: Pre-Processing User-Invocable: false

Purpose

Parse and normalize Jira issue fields into a structured format for STP generation.

When to Use

Invoked by the jira-collector subagent after fetching raw Jira issue data.

Input

Raw Jira issue response containing:

• Standard fields (summary, description, status, priority, etc.)
• Custom fields (Feature Link, Git Pull Request, etc.)
• Issue links (outward and inward)
• Comments

Output Format

parsed_issue:
  key: PROJ-12345
  summary: <issue summary>
  description: <full description text>
  status: <In Progress | Done | etc>
  issue_type: <Story | Bug | Enhancement | Epic | Task>
  priority: <Critical | Major | Minor | etc>
  labels:
    - label1
    - label2
  components:
    - component1
    - component2
  acceptance_criteria: <extracted from description or custom field>
  feature_link: <URL or null>
  git_pull_requests:
    - https://github.com/...
  reporter:
    name: <name>
    email: <email>
  assignee:
    name: <name>
    email: <email>
  created: <date>  # ISO 8601 format
  updated: <date>  # ISO 8601 format
  fix_version: <version string or null>
  custom_fields:
    - id: customfield_12345
      name: Feature Link
      value: <value>
    - id: customfield_67890
      name: Git Pull Request
      value: <value>
    - ...  # All non-null custom fields

Field Extraction Rules

Standard Fields

Jira Field	Output Field	Notes
summary	summary	Direct mapping
description	description	Full text, may contain markdown
status.name	status	Status display name
issuetype.name	issue_type	Issue type display name
priority.name	priority	Priority display name
labels	labels	Array of strings
components[].name	components	Array of component names
fixVersions[0].name	fix_version	First fix version or null
reporter	reporter	{name, email} object
assignee	assignee	{name, email} object
created	created	ISO 8601 timestamp
updated	updated	ISO 8601 timestamp

Custom Fields

Custom field IDs vary by Jira instance. Search for these patterns:

Custom Field Name	Common Patterns	Output Field
Feature Link	customfield_* containing "Feature", "Epic Link"	feature_link
Git Pull Request	customfield_* containing "Git", "Pull Request", "PR"	git_pull_requests

Custom Field Extraction Rules

Extract ALL non-null custom fields into the custom_fields array:

Rule	Description
Include	All custom fields with non-null, non-empty values
Exclude	System fields (already mapped above)
Format	{id, name, value} for each field
Name Discovery	Use Jira API field metadata to get display names
Value Types	Strings, arrays, objects - preserve original structure

MANDATORY Custom Fields to Look For:

1. Git Pull Request - Contains PR URLs (critical for PR collection)
2. Feature Link / Epic Link - Parent feature reference
3. Acceptance Criteria - Testing requirements
4. Story Points - Estimation data
5. Sprint - Sprint assignment
6. Target Version - Release targeting

Acceptance Criteria Extraction

Look for acceptance criteria in:

1. Custom field named "Acceptance Criteria"
2. Description section starting with "Acceptance Criteria:" or "AC:"
3. Description section with h3. Acceptance Criteria (Jira wiki markup)

Comment Processing

For each comment, extract:

comments:
  - id: <comment id>
    author:
      name: <display name>
      email: <email if available>
    created: <timestamp>
    body: <comment body>
    pr_urls_found:
      - <any GitHub PR URLs found in body>

GitHub PR URL Extraction

Scan text for patterns matching:

• https://github.com/{owner}/{repo}/pull/{number}
• github.com/{owner}/{repo}/pull/{number} (without https)

Return deduplicated list of full URLs.

Normalization Rules

1. Text cleaning: Remove excessive whitespace, normalize line endings
2. URL normalization: Ensure all URLs have https:// prefix
3. Empty handling: Use null for missing optional fields, empty arrays for lists
4. Date formatting: ISO 8601 format (YYYY-MM-DDTHH:MM:SSZ)

Example

Input (raw Jira response):

{
  "key": "PROJ-12345",
  "fields": {
    "summary": "Add hot-plug support for CPU",
    "description": "As a user, I want to...\n\nAcceptance Criteria:\n- CPU can be added\n- CPU can be removed",
    "status": {"name": "In Progress"},
    "issuetype": {"name": "Story"},
    "priority": {"name": "Major"},
    "labels": ["hot-plug", "cpu"],
    "components": [{"name": "app-controller"}],
    "customfield_12345": "https://your-jira.example.com/browse/PROJ-99000",
    "customfield_67890": "https://github.com/example-org/example-repo/pull/1234"
  }
}

Output:

parsed_issue:
  key: PROJ-12345
  summary: Add hot-plug support for CPU
  description: "As a user, I want to...\n\nAcceptance Criteria:\n- CPU can be added\n- CPU can be removed"
  status: In Progress
  issue_type: Story
  priority: Major
  labels:
    - hot-plug
    - cpu
  components:
    - app-controller
  acceptance_criteria: |
    - CPU can be added
    - CPU can be removed
  feature_link: https://your-jira.example.com/browse/PROJ-99000
  git_pull_requests:
    - https://github.com/example-org/example-repo/pull/1234