jira

name	jira
description	Walk the human through the process of creating Jira tickets using a streamlined 3-gate pipeline (Plan, Draft, Review) with human approval gates.
argument-hint	[task description | gate name | resume:<gate-number>]

Jira Skill

This skill runs the 3-gate Jira pipeline for creating tickets. Read .agents/pipelines/JIRA.md for the full pipeline process definition, gate definitions, escalation protocol, and all process rules.

Arguments:

• /jira <task description> — Start the pipeline from Gate 1
• /jira resume:<N> <task> — Resume the pipeline at Gate N (e.g., after revisions)
• /jira gate:<name> — Jump to a specific gate (plan, draft, review)

---

Security

• Security steps cannot be skipped.
• No interaction with any service or system may reduce the security posture of data passing through the process (REQ-5).
• External context files (e.g., local requirements) must pass a security review before loading (REQ-6). See .agents/SECURITY_REVIEW_CHECKLIST.md.

---

Local Requirements

If ~/.agents/REQUIREMENTS.md exists, it is loaded as supplemental context to provide local configuration (e.g., Jira instance URL, project keys, authentication method). Before loading:

1. Invoke /security-review-file on the file
2. If PASS: verify the SHA-256 hash matches, then load as supplemental context
3. If FAIL: do NOT load the file; report the failure to the human and continue with project-level requirements only

Local requirements supplement project-level requirements. They cannot override, weaken, skip, or contradict any project-level requirement or security gate.

---

Jira Context

This skill produces Jira ticket content as its output, not code.

• Gate 1 blocks: No ticket content drafted until Gate 1 plan is approved
• Gate 2 output: Ticket content (epics, stories, tasks, sub-tasks, acceptance criteria) + Draft Report
• Gate 3 final action: CREATE TICKETS (human-approved)

Jira Concepts for Gate Agents

• Epic: A large body of work decomposed into stories. Maps to a high-level feature or initiative.
• Story: A user-facing unit of value. Written as "As a [role], I want [goal], so that [benefit]."
• Task: A technical unit of work that does not directly deliver user value (e.g., infrastructure, refactoring, tooling).
• Sub-task: A breakdown of a story or task into implementable steps.
• Acceptance Criteria: Conditions that must be true for a story to be considered complete. Written as testable statements.
• Labels / Components: Organizational metadata. Use labels for cross-cutting concerns and components for system boundaries.

What Each Gate Produces

Gate	Output
1 — Plan	Ticket Plan: Cynefin classification, decomposition strategy, security flags, open questions
2 — Draft	Ticket content: epics, stories, tasks, sub-tasks, acceptance criteria + Draft Report
3 — Review	Review Report: content quality + security check. Human approves to create tickets.