SKILL.md
Launch Tier Planner
Decides how big a launch is and what kind it is — the tier (Tier 1 flagship all-channel / Tier 2 targeted / Tier 3 changelog-level), the type (new-product / feature / relaunch / partnership), the effort that tier justifies, the KPI targets declared before launch, and the risk register with kill criteria that the day-of runbook inherits. It sits in the Research phase of the RAMP loop and feeds the RAMP R sub-items launch tier & type declared with effort calibrated, risk register exists (likelihood × blast-radius, owners, kill criteria / rollback thresholds), and launch KPI targets (D0/W1/M1) declared before launch. Sizing the moment correctly is what keeps a changelog entry from burning a Tier-1 audience and a flagship from shipping with a Tier-3 kit.
Scope guard: this skill sizes the launch and registers its risks only. It does not pick the date or window (that is launch-window-planner), build the positioning canvas (that is positioning-mapper), run a creator-channel launch campaign (launch requests that mention creators route to campaign-planner), compute the RAMP profile result or run the RAMP vetoes (launch-readiness-auditor), or write stage/date/tier facts to memory/launch-registry/ directly (launch-registry is the sole writer — this skill submits candidates). It works one lever — sizing — and hands off.
Quick Start
How big should the launch of [product / feature] be? Audience: [who is affected]. Revenue link: [direct / indirect / none].
Declare tier and type for [launch], build the risk register with kill criteria, and sketch the T-8w to T+4w timeline.
This is a partnership launch with [partner] — set the tier, split the co-marketing responsibilities, and set D0/W1/M1 targets.