product-roadmap-focus

license	FSL-1.1-MIT
name	product-roadmap-focus
description	Chooses the most important product or roadmap work under ADHD, founder, budget, and multi-agent constraints. Use when asking what to work on next, prioritizing a roadmap, cutting scope, choosing an MVP slice, or deciding whether to split work across agents. NOT for executing an already-chosen task, generic project status reporting, personal financial advice, or long unbounded brainstorming.
allowed-tools	Read,Write,Edit,Bash,Grep,Glob
metadata	{"category":"Product & Strategy","tags":["product-focus","roadmap","prioritization","adhd","founder","agent-coordination"],"pairs-with":[{"skill":"project-management-guru-adhd","reason":"Converts selected roadmap work into ADHD-friendly chunks and reminders."},{"skill":"tech-entrepreneur-coach-adhd","reason":"Applies MVP, money, and scope tests before building."},{"skill":"cqrs-event-sourcing-architect","reason":"Records roadmap decisions as append-only events and projections."}],"provenance":{"kind":"first-party","owners":["port-daddy"]},"mirrors":{"repo":"skills/product-roadmap-focus","codex":".codex/skills/product-roadmap-focus","claude":".claude/skills/product-roadmap-focus","agents":".agents/skills/product-roadmap-focus"}}
io-contract	{"kind":"deliverable","produces":[{"kind":"design-doc","description":"Prioritized roadmap slice with scope cuts, MVP selection, and explicit deferral rationale under founder/budget constraints"}]}

Product Roadmap Focus

Use this skill when everything feels important. Its job is to force one clear roadmap commitment, name what is being cut, and leave evidence that future agents can replay instead of re-litigating the same choice.

This is not a generic backlog grooming skill. It is for ADHD founder/product work where technical excitement, operator frustration, money pressure, and multi-agent sprawl all compete for attention.

Core Rule

The best next task is the smallest visible slice that most increases the odds of a real user choosing the product again.

If a candidate does not improve operator trust, reduce a current blocker, create customer proof, reduce existential risk, or unlock many follow-on tasks, it is probably not "now." It may still be good. Put it in not-now with dignity.

Decision Flow

flowchart TD
  A["Start: too many possible next moves"] --> B["Archaeology: what is already true?"]
  B --> C["Constraints: runway, deadline, trust, context, active PRs"]
  C --> D["Candidate inventory: 5-12 options max"]
  D --> E{"Any urgent trust or broken-core issue?"}
  E -->|Yes| F["Pick the smallest trust-restoring slice"]
  E -->|No| G{"Any slice can create customer/operator proof this week?"}
  G -->|Yes| H["Pick the proof slice"]
  G -->|No| I{"Any technical unlock gates many other slices?"}
  I -->|Yes| J["Pick the unlock slice"]
  I -->|No| K["Pick the highest learning-per-hour slice"]
  F --> L["Write focus receipt"]
  H --> L
  J --> L
  K --> L
  L --> M["Chunk into 30-90 minute steps or a small DAG"]
  M --> N["Record decision event and not-now list"]

The Seven Lenses

Score lightly. Do not turn this into spreadsheet theater. Use 0, 1, 2, 3.

Lens	Ask	High score means
Operator trust	What is currently making the product feel fake, scary, or unusable?	Fixing it makes the product credible today.
Customer proof	Does this produce screenshots, transcripts, demos, revenue, or user feedback?	A human can react to it this week.
Money/risk	Does this reduce burn, avoid waste, or improve the path to paid use?	It changes survival odds, not vanity.
Technical unlock	Does this unblock several other slices?	Many future tasks get cheaper.
Context economy	Does this reduce repeated archaeology or context loss?	Future agents stop relearning the same thing.
ADHD fit	Can this be finished in a few visible chunks without shame spirals?