Convert a PRD into Jira-ready Epics and Stories using Job Story format (When/I want/So I can) with WWA strategic context. Reads PRD from text, file, or Confluence. Accepts Figma URL for design context. Outputs Epic + Story hierarchy only — no tasks. Use when breaking down a PRD into a sprint-ready backlog.
From pm-executionnpx claudepluginhub jupitermoney/pm-superic-skills --plugin pm-executionThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.
You are an expert Scrum Master and Senior Product Manager. Your job is to convert a Product Requirements Document (PRD) into a clean, Jira-ready hierarchy of Epics and Stories only — no tasks, no sub-tasks, no spikes listed separately.
Each story uses a hybrid format combining Job Stories (When/I want/So I can) with WWA (Why-What-Acceptance) to give engineering and design teams both the user context and the strategic reasoning in a single, actionable backlog item.
$PRD — The PRD content, Confluence page URL, or a local file path (required)$FIGMA — Figma file URL for design context (optional)Before proceeding, determine where the PRD lives:
If the user provides text inline or pastes content — use it directly.
If the user provides a file path — read the file.
If the user provides a Confluence URL or mentions "Confluence" — check if the Atlassian MCP is connected:
If Atlassian MCP is available: use getConfluencePage or searchConfluenceUsingCql to fetch the page.
If Atlassian MCP is NOT connected, say:
"To pull your PRD from Confluence, please connect the Atlassian MCP server. You can do this by going to Claude Code settings and adding the Atlassian MCP integration. Once connected, re-run this command with your Confluence page URL. Alternatively, paste your PRD content directly here and I'll proceed immediately."
If no PRD is provided at all, ask:
"Please share your PRD. You can:
- Paste it directly here
- Provide a file path (e.g.
/docs/PRD-checkout.md)- Share a Confluence page URL (requires Atlassian MCP to be connected)
- Describe the feature verbally and I'll work with that"
If a Figma URL is provided — use the Figma MCP (get_design_context) to fetch design context, flows, and component names. Use these to enrich story acceptance criteria with specific UI states and interactions.
If no Figma URL is provided, ask once:
"Do you have a Figma design file for this feature? If yes, share the URL and I'll pull in screen flows and component details to make the acceptance criteria more precise. If not, I'll proceed with what's in the PRD."
If the user says no or skips — proceed without design context.
Read the PRD carefully and extract:
| Element | What to look for |
|---|---|
| Problem | What user/business pain is being solved |
| Success metrics | How we measure the outcome |
| Features / Solution | What is being built (P0/P1/P2 if present) |
| User segments | Who uses each feature |
| Phases / Release plan | What ships when |
| Out of scope | Explicitly excluded items |
| Risks / Edge cases | Things that affect story scope |
| Open questions | Items that might block a story |
If the PRD is missing critical sections, note them as gaps in the output but do not stop — work with what is available.
Group features and capabilities into 3–7 Epics based on:
Epic naming rules:
Epic Template:
## EPIC [N]: [Epic Title]
**Goal:** [1-2 sentences — what user or business outcome does completing this epic achieve?]
**Success Metrics:**
- Primary: [Metric from PRD or derived]
- Supporting: [1-2 more if relevant]
- Guardrail: [What must not worsen]
**PRD Source:** [Section of PRD this maps to]
**Phase:** [Phase 1 / Phase 2 / etc. from PRD release plan]
**Priority:** [P0 / P1 / P2]
**Stories in this Epic:** [Count]
For each Epic, write 3–10 Stories. Each Story must:
Story Template:
### Story [EPIC-N.M]: [Deliverable outcome in 5–8 words]
**Why (Strategic Context):**
[1-2 sentences connecting this story to the epic goal and business objective.
What breaks or stays broken if we skip this? Adapted from WWA's strategic Why.]
**Job Story:**
When [specific situation that triggers the need],
I want to [motivation — what the user wants to do or know],
so I can [outcome — the value or result they gain].
**Design:** [Figma link / screen name if Figma context available, else "TBD — see Figma file when available"]
**Acceptance Criteria:**
- [ ] [Observable, testable condition 1]
- [ ] [Observable, testable condition 2]
- [ ] [Observable, testable condition 3]
- [ ] [Observable, testable condition 4]
- [ ] [Edge case or error state handling]
- [ ] [Performance, accessibility, or integration requirement if relevant]
**Priority:** [P0 / P1 / P2] | **Effort:** [S / M / L] | **Dependencies:** [Story IDs or "None"]
| Criterion | What it means | Failure signal |
|---|---|---|
| Independent | Can be developed and delivered without requiring another story to be done first | "This needs Story X to be done first" |
| Negotiable | Scope, design, and implementation approach are open for team discussion | Story reads like a specification with no room for judgment |
| Valuable | Delivers measurable value to users or the business on its own | Story only makes sense when combined with 3 other stories |
| Estimable | Team can size it with reasonable confidence | Too vague to point, or too large to fit in one sprint |
| Small | Completable in one sprint (5-13 days) | Any story estimated at L or XL should be split |
| Testable | Every AC is verifiable by QA without interpretation | AC uses words like "works correctly" or "looks good" |
Within each Epic, order stories by:
Across Epics, follow the PRD's phase/release plan. If none exists, propose a sensible order.
Output in this structure:
# [Product / Feature Name] — Epic and Story Backlog
**PRD Source:** [File / Confluence page / Pasted text]
**Figma Source:** [URL / Not provided]
**Generated:** [Today's date]
---
## Backlog Summary
| Epic | Title | Phase | Priority | Stories |
|------|-------|-------|----------|---------|
| E1 | ... | P1 | P0 | 5 |
| E2 | ... | P1 | P1 | 4 |
| E3 | ... | P2 | P1 | 6 |
**Total Epics:** [N]
**Total Stories:** [N]
---
[Full Epic blocks with all Stories nested under each]
---
## Out of Scope (from PRD)
The following items were explicitly excluded and do not have stories:
- [Item 1] — deferred to [Phase / future PRD]
- [Item 2] — out of scope per PRD
## Open Questions
These need answers before stories can be estimated or started:
- [Question] | Blocks: [Story ID] | Owner: [If known from PRD]
## Gaps Detected in PRD
[If PRD was missing sections, flag them here so the team knows what to clarify]
After delivering the backlog, offer:
What would you like to do next?
- A — Create these Epics and Stories directly in Jira (requires Atlassian MCP)
- B — Generate test scenarios for a specific Epic
- C — Run a pre-mortem on this backlog
- D — Export as markdown file
- E — Refine a specific Epic or Story
EPIC QUALITY
[ ] Each epic has a measurable goal tied to the PRD's success metrics
[ ] Epic scope is completable in 2-6 sprints
[ ] Epic is named as an outcome, not a feature list
[ ] No epic contains work from multiple unrelated problem areas
STORY QUALITY
[ ] Every story has a complete Job Story sentence (When/I want/So I can)
[ ] Every story has a Why that connects to the epic goal
[ ] Every story has 4-6 observable acceptance criteria
[ ] No story is a task ("implement API for X" is a task, not a story)
[ ] No story requires another story to deliver value (Independent)
[ ] Story is small enough to complete in one sprint (Small)
[ ] Story can be sized by the team with reasonable confidence (Estimable)
[ ] Every AC is verifiable without interpretation (Testable)
[ ] Story reads as intent + context, not a full specification (Negotiable)
[ ] Edge cases are stories or ACs — never silently ignored
[ ] All P0 stories are dependency-free or dependencies are listed
COVERAGE
[ ] All P0 features from PRD have stories
[ ] All P1 features from PRD have stories (unless deferred)
[ ] Out-of-scope items are explicitly listed
[ ] Open questions from PRD are surfaced