Menu
Quarterly portfolio review. Hard call on what to kill, what to maintain, what to double down on. Forces decisions.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Slice a goal into the smallest feature slices that can each be PROVEN — every slice ships with a named eval and an end-to-end test, and a slice you can't evaluate or test doesn't get selected. Use after the goal is set and before the spec, so only provable features make the build.
Guided, stateful provisioning of a product's production stack — deep-links, key-format validation, a secrets backend (fnox recommended, or user-touched .env.local), a resumable ledger, CLI automation after signup, and a blocking A1–A10 live gate. Hands off to /security-check → /ship. The walked version of SETUP.md.
Creates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when requirements are unclear, ambiguous, or only exist as a vague idea.
Launch the factory's control plane — scaffold and fill FACTORY-ORDERS (weekly mandate + hard budget), STANDING-ORDERS (autonomous-program authority), HEARTBEAT (weekly pulse), and write lifecycle gates for every registered product. A deliberate one-time ritual (re-run to audit); invoke as /factory-launch, not auto-triggered.
One-shot a new product — folders, Next.js starter, product.config.json, CLAUDE.md, scope.md, PRD skeleton, SETUP.md checklist. The Lovable-like ignition.
Breaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.
| name | kill-or-double-down |
| description | Quarterly portfolio review. Hard call on what to kill, what to maintain, what to double down on. Forces decisions. |
| model_tier | opus |
/kill-or-keep skillSaved to brain/decision-log/portfolio-review-YYYY-QN.md:
## Portfolio Review — <Q-YYYY>
### Snapshot
| Product | Stage | MRR | Trend | Founder hours/wk | Verdict |
|---|---|---|---|---|---|
| linkedup | mvp | $0 | flat | 2 | <verdict> |
| ventbox | launch | $200 | +20% | 4 | <verdict> |
Verdict legend: **DOUBLE-DOWN** | **MAINTAIN** | **MINIMAL** | **KILL**
### DOUBLE-DOWN (allocate the most hours next quarter)
- **<product>**: <why> — proposed quarterly target: <specific metric>
- Top 3 initiatives:
1. ...
2. ...
3. ...
### MAINTAIN (steady hand, no major investment)
- **<product>**: keep shipping, monitor metrics, no new feature work
- Top recurring tasks: <list>
### MINIMAL (life support — < 2 hours/week)
- **<product>**: keep alive, do not invest, revisit in 3 months
- Why not killed: <reason — e.g. SEO compounding, low maintenance>
### KILL (sunset)
- **<product>**: <reason — usually: no PMF after N months, no clear path>
- Sunset plan:
1. Email users (T+0): announce, give 60 days
2. Disable signups (T+0)
3. Export data for any paying customer (T+30)
4. Shut down (T+60)
5. Archive code (T+60)
6. Write learnings to `brain/decision-log/postmortem-<product>-YYYY-MM-DD.md`
### Portfolio-level decisions
1. <decision — e.g. "stop scaffolding new products this quarter, focus on the 2 hitting PMF">
2. <decision — e.g. "raise prices on Ventbox by 30%">
3. <decision — e.g. "kill the calculatrs slot if no movement by EOQ">
### What I'm getting wrong
A self-check: where am I likely being too optimistic or too pessimistic? <one paragraph>
product.config.json, metrics.md, last 3 months of decisions/.Fleet mode (when subagent spawning is available): per factory/playbooks/mvp-stage/fleet-patterns.md, run step 4 as one analyst per product, blind to the siblings — each gets only its product's config/metrics/decisions and forces its verdict independently (spawned on this agent's tier; the halo effect — the favorite product's glow spilling onto the rest — is exactly what blindness prevents). Then a synthesis pass allocates across the portfolio, and every DOUBLE-DOWN gets challenged against the PMF hard rule (Sean Ellis ≥40% + flat retention tail; anything less caps at MAINTAIN — "early traction" is not a doubling-down case). Serial remains valid for a quick pass.
Killing a product is one of the highest-leverage decisions a portfolio founder can make. Hours not spent on a dead product compound on the live ones. Common reluctance: sunk cost. Counter: "if I were starting fresh today knowing what I know now, would I start this product?" If no → kill.
metrics.md, decisions/, product.config.jsonbrain/operating-principles.mdfactory/playbooks/founders-wisdom/100k-arr-tactics.md