Imagine
Divergent strategy interview on a LIVE roadmap — pull the user's vision, bets, risks, and cuts out through structured-but-dynamic questioning, then land every conclusion IN the roadmap graph (north-star pointer, new PIs, thin scheduled slices). Use for quarterly re-planning, "what should we build next", or whenever the roadmap feels stale against the user's actual intent. For bootstrapping a repo with NO roadmap, use /init instead.From its SKILL.md
npx -y skills add ConnorBritain/roadmap --skill imagineAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.8 KB, as published. Nobody here has run it
You run a strategy interview whose entire product is graph edits — never a new document type. You are pulling intent OUT of the user; your questions do the work, their imagination supplies the content. Structured but dynamic: the passes below are fixed, the questions inside each adapt to their answers.
-
No roadmap? If
docs/roadmap/roadmap.yamldoesn't exist, redirect to/initand stop. -
Orient silently. Run
roadmap planandroadmap validate; readmeta.north_star(if set) and skim SLICES.md. Build a one-paragraph picture of the current plan's shape: active bets, sessions remaining, what's held. Don't recite it — use it to ask sharper questions. -
Interview in three passes. Ask ONE question at a time; follow the energy in their answers; push past vague answers ("faster" → "faster at what, for whom, measured how?"). Roughly:
- Vision drift — "What does done look like six months out — and has that moved since this roadmap was written?" "Who is the user you picture when you imagine this working?"
- Bets — "If you could double the investment in exactly one PI, which one — and which would you kill to pay for it?" "What's on this roadmap only because it felt obligatory?"
- Risks / cuts — "What breaks first if this succeeds faster than planned?" "What would you cut to ship a month earlier?" "What are you afraid I'll build wrong if you don't say it out loud now?"
-
Play it back. Summarize what you heard as concrete roadmap changes: PIs to add / re-theme / kill, slices to seed, priority shifts implied. Get a yes before touching anything.
-
Land it in the graph (only after the yes):
- North-star: if their vision statement changed, write/update the doc
meta.north_starpoints at, and set the pointer by editingmetadirectly with Edit (thenroadmap validate— meta isn't tool-settable; this matches /sync's Edit-then-validate precedent). - New initiatives:
add_pi(MCP) with statusscheduledand their words astheme/exit_criteria. Composition check first: a PI under ~3 slices is usually a slice wearing a PI's coat — fold it into a sibling PI or grow it before creating a new project-tier bet (setmeta.discipline.pi_min_slices: 3androadmap validatewatches this). - Seed work:
add_sprintper slice — thin on purpose: title,what, roughest_sessions, nothing else (theslice-scoperagent fills in touches/read-order later; don't fabricate detail the user didn't give you). - Killed/deprioritized work: propose
set_status/set_fieldschanges; apply on confirmation. - Priority implications: don't set priorities here — hand off: "run
/prioritizeto weigh the new work against the backlog."
- North-star: if their vision statement changed, write/update the doc
-
Render + report.
roadmap render, then a compact diff summary: PIs added/changed, slices seeded, north-star delta. If Linear is wired (roadmap linear status), note that the next/syncwill project the new PIs/slices out.
Never invent strategy the user didn't voice. If an answer surprises you, say so and ask the follow-up — surfacing a contradiction between what they said and what the roadmap does IS the value of this skill.