$npx -y skills add Productfculty-aipm/PM-Copilot-by-Product-Faculty --skill epic-breakdownUse this skill when the user asks to "break this initiative into epics", "structure this as epics", "how should I organize this work", "help me sequence this initiative", "epic planning", "what are the phases of this project", "initiative breakdown", or has a large initiative and
| 1 | # Epic Breakdown |
| 2 | |
| 3 | You are helping the user decompose a large initiative into a sequenced set of epics — each of which delivers user value independently and can be shipped and measured before the next one begins. |
| 4 | |
| 5 | Framework: Shape Up (appetite-based scoping), Lenny's PRD guide (just enough direction), iterative delivery principles. |
| 6 | |
| 7 | ## Step 1 — Load Context |
| 8 | |
| 9 | Read `memory/user-profile.md` for team size, current roadmap state, and any existing epics or initiatives in the Now/Next list. Read `context/product/roadmap.md` for OKR context. |
| 10 | |
| 11 | ## Step 2 — Understand the Initiative |
| 12 | |
| 13 | Extract from the user's description: |
| 14 | - What is the full initiative (if shipped completely, what does it look like)? |
| 15 | - What user problem does it solve? |
| 16 | - What's the constraint: time, team size, or scope? |
| 17 | - What would a "minimal first ship" look like? |
| 18 | |
| 19 | ## Step 3 — Identify Natural Seams |
| 20 | |
| 21 | Find the natural breaking points in the initiative — places where a partial version delivers real user value. These become epic boundaries. |
| 22 | |
| 23 | Ask: "If we shipped only this piece, would users get real value? Or would it be invisible to them?" |
| 24 | |
| 25 | Good epic breaks: |
| 26 | - Complete user flow (even if limited) that a user can accomplish end-to-end |
| 27 | - A new user segment being served for the first time |
| 28 | - A new connection to an external tool that stands alone |
| 29 | - A new mode or capability that can be toggled independently |
| 30 | |
| 31 | Bad epic breaks: |
| 32 | - Half a user flow (user can start but not finish) |
| 33 | - Backend work with no user-visible output |
| 34 | - A step that's only valuable in combination with the next step |
| 35 | |
| 36 | ## Step 4 — Sequence the Epics |
| 37 | |
| 38 | Order epics using these principles: |
| 39 | - **De-risk first:** Tackle the epic with the most technical or user uncertainty first. Learn before you invest. |
| 40 | - **Value early:** If possible, ship something users can use and give feedback on before the full initiative is complete. |
| 41 | - **Dependencies first:** If Epic 2 depends on Epic 1's infrastructure, sequence accordingly. |
| 42 | - **80/20 rule:** The first epic should deliver 80% of the user value in 20% of the work. |
| 43 | |
| 44 | ## Step 5 — Epic Template |
| 45 | |
| 46 | For each epic, fill in: |
| 47 | |
| 48 | **Epic name:** [3–5 words, what it does] |
| 49 | **User value:** [One sentence — what can users do after this ships that they couldn't before?] |
| 50 | **Scope:** [What's included; what's explicitly excluded] |
| 51 | **Success metric:** [How will we know this epic delivered value?] |
| 52 | **Size estimate:** [Number of sprints / person-weeks — be explicit about assumptions] |
| 53 | **Dependencies:** [What must be done before this epic can start?] |
| 54 | **Risks:** [What could cause this epic to take 2× longer?] |
| 55 | **MVP vs. Full:** [Is this epic the MVP version or the full version?] |
| 56 | |
| 57 | ## Step 6 — Output |
| 58 | |
| 59 | Produce: |
| 60 | - Epic breakdown table (name, user value, size, sequence, status) |
| 61 | - Sequencing rationale (why in this order?) |
| 62 | - Ship plan: which epics can go live incrementally vs. need to wait for others |
| 63 | - Recommendation for Epic 1 starting point (the minimum shippable first piece) |
| 64 | - A question to validate sequencing: "Would you rather ship [Epic 1 value] in [Epic 1 time] and then [Epic 2 value], or wait to ship everything together? Why?" |