From MARKET-SCANNER — Discover what's worth building
Set or refine the standing DISCOVERY GOAL that scans run over. Trigger with /discovery-goal (or "set a discovery goal", "I want to find a SaaS in <niche>", "refine the goal", "loosen the constraints"). Captures the constraints on WHAT to discover (niche, builder edge, target price band, stack-fit, effort appetite, hard constraints) — infer-first, one question at a time with a recommended answer + multiple-choice — and writes them to .market-scanner/goal.md so /market-scan and /loop scan a bounded space.
How this skill is triggered — by the user, by Claude, or both
Slash command
/market-scanner:goal-setterinheritThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A `/discovery-goal` bounds the search so scans are focused, not infinite — the constraints on *what to discover*,
A /discovery-goal bounds the search so scans are focused, not infinite — the constraints on what to discover,
distinct from an IDEA (one chosen thing to build). The full field set and the loop it drives live in
../../knowledge/discovery/goal-loop.md.
.market-scanner/goal.md if present (refining, not starting fresh)..market-scanner/goal.md (create .market-scanner/ if absent) and confirm the goal back to the
user in a compact block./market-scan once, or /loop /market-scan to iterate over the goal until a
candidate earns a keep verdict.Keep the goal tight enough to focus, loose enough to surprise. Over-constrained goals starve the scan (only already-killed shapes appear); under-constrained goals scatter it. When a loop keeps returning killed shapes, the fix is usually here — loosen one constraint or shift the niche.
Carries the KAIZEN self-improvement covenant (../../knowledge/covenant.md).
npx claudepluginhub agentic-underground/idea-to-productionProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.