Skip to main content

wild-to-mild

Generate ideas at deliberate points along the originality-feasibility spectrum, from moonshots to Monday-morning moves. Use when the user needs to see the full range of what's possible — not just "good ideas" but the actual shape of the option space from wild to safe. Triggers on "give me a range," "what are my options from conservative to aggressive," "I want to see the whole spectrum," "what's the bold version and what's the safe version," "give me the crazy idea and the realistic one," or any moment where the user needs to make a strategic choice about HOW ambitious to be, not just WHAT to do. Also use when every idea in a brainstorm clusters at the same risk level — all safe, or all moonshots — and the user can't see the tradeoffs because there's nothing to compare against. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. This skill doesn't pick a point on the spectrum. It shows you the whole line and lets you choose.

설치로 이동

소스 정보

저장소
d4vidc4rson/divergent-agents
최근 소스 활동
2026년 4월 1일 12:58
감지된 SKILL.md 언어
영어
스타
1
포크
0

설치 방법

기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.

소스 파일 검토

설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.

파일 탐색기
3 개 파일

SKILL.md 표시 중

SKILL.md
소스 지침 · 읽기 전용 미리보기
name
wild-to-mild
description
Generate ideas at deliberate points along the originality-feasibility spectrum, from moonshots to Monday-morning moves. Use when the user needs to see the full range of what's possible — not just "good ideas" but the actual shape of the option space from wild to safe. Triggers on "give me a range," "what are my options from conservative to aggressive," "I want to see the whole spectrum," "what's the bold version and what's the safe version," "give me the crazy idea and the realistic one," or any moment where the user needs to make a strategic choice about HOW ambitious to be, not just WHAT to do. Also use when every idea in a brainstorm clusters at the same risk level — all safe, or all moonshots — and the user can't see the tradeoffs because there's nothing to compare against. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. This skill doesn't pick a point on the spectrum. It shows you the whole line and lets you choose.
# Wild to Mild Every brainstorm has a default altitude. Most fly at cruising altitude: ideas that are interesting enough to feel creative but safe enough to not get anyone fired. The middle of the spectrum. The place where most ideas live and most ideas die because they're not bold enough to matter and not practical enough to ship tomorrow. The LiveIdeaBench research (Ruan et al., 2026) found a Pareto front between originality and feasibility. The most original ideas tend to be the least implementable. The most implementable ideas tend to be the least original. This isn't a flaw. It's the actual shape of the option space. And most brainstorms hide it from you by only producing ideas at one altitude. This happens in every domain. Business brainstorms cluster at "ambitious but fundable." Engineering brainstorms cluster at "doable in a sprint." Research brainstorms cluster at "publishable." Policy brainstorms cluster at "politically viable." Education brainstorms cluster at "within existing curriculum." Each domain has a cruising altitude it defaults to. This skill shows you the whole range — from the thing you could do before lunch to the thing that might not be possible but would change everything if it were. Then you pick your altitude. --- ## How It Works Generate ideas at four deliberate points on the spectrum. Not mixed together. Separated. Each altitude is its own section with its own rules. ### Altitude 1: Monday Morning Ideas you could execute this week with the resources you already have. No new hires. No new budget. No new tools. No new funding. No new approvals. No permission needed from anyone you don't already talk to. These ideas are not exciting. That's the point. They're valuable because they're real. They can happen while the bigger ideas are still being debated. The question at this altitude: **"What could I do before Friday that would move this forward?"** Rules: - Must be completable in under a week - Must require zero new resources - Must be something one person could own - Can be ugly, scrappy, or temporary - Does not need to scale Generate 3-4 ideas at this altitude. ### Altitude 2: This Quarter Ideas that require some effort, planning, or investment but are clearly achievable within 90 days. Might need a small budget. Might need one new hire, collaborator, or partnership. Might need buy-in from a boss, board, or committee. But nothing here is speculative. You know how to do these things. The question at this altitude: **"What could we commit to this quarter and be confident it ships?"** Rules: - Must be achievable in 90 days - Can require modest new resources - Should have a clear owner and clear milestones - Should produce a measurable result (not "build awareness" or "explore the space" — something you can point to and say "this happened") Generate 3-4 ideas at this altitude. ### Altitude 3: Stretch Ideas that would require significant new capability, investment, or risk. You don't know exactly how to do these yet. They'd take 6-12 months. They might fail. But if they work, they meaningfully change the trajectory of the thing. The question at this altitude: **"What would we do if we were twice as brave as we currently are?"** Rules: - Must require something you don't currently have (skill, money, partnership, technology, permission, political capital, institutional support) - Must have a credible path even if it's uncertain - Should make at least one stakeholder nervous - For each idea, state the one condition that would need to be true for it to work: "This works if ___" Generate 3-4 ideas at this altitude. ### Altitude 4: Moonshot Ideas that might not be possible. You don't know if the physics work. You don't know if the politics allow it. You don't know if anyone would say yes. You don't know if the technology exists. But if this worked, it would be the thing people talk about for years. The question at this altitude: **"What would we do if failure had no consequences?"** Rules: - Must make a reasonable person in the domain say "that's insane" or "you can't do that" - Must be genuinely transformative, not just big (a bigger version of a safe idea is not a moonshot — a fundamentally different approach is) - Does not need a credible path - For each idea, state what it would change if it worked: "If this works, ___" Generate 2-3 ideas at this altitude. --- ## Why Four Altitudes, Not Two Most people think in two modes: the safe thing and the big thing. "We could do X, or we could go big and do Y." Two options is a coin flip, not a strategy. You can't see the tradeoff curve with only two points on it. Four altitudes show you the actual shape: - What's available right now for free - What's available with modest investment - What requires real risk - What requires a leap of faith The space between Altitude 2 and Altitude 3 is where the most interesting strategic decisions live. That's the border between "we know how to do this" and "we'd have to figure it out." In business, it's the border between optimization and reinvention. In science, it's the border between incremental experiments and paradigm-challenging ones. In policy, it's the border between reform and transformation. In engineering, it's the border between refactoring and rewriting. Most people never see that border clearly because they either stay below it or fantasize above it. --- ## After Generation: The Tradeoff Conversation Once you've generated ideas at all four altitudes, present them grouped by altitude. Then surface the tradeoff explicitly: **What you get by staying low (Altitude 1-2):** Speed. Certainty. Low risk. Momentum now. But: the ideas at this altitude probably don't change the game. They improve the current trajectory. They don't create a new one. **What you get by going high (Altitude 3-4):** Differentiation. Transformation. The ideas that competitors, rival labs, or opposing factions can't copy because they're too weird or too hard. But: they take longer, cost more, and might not work. **The real question:** "Given where you are right now, which altitude should get most of your energy? And is there a Monday Morning idea you should start immediately while the bigger bets are being built?" Almost always, the right answer is a Monday Morning idea running now plus one Stretch idea being developed. But that's a portfolio decision only the user can make. Name the tradeoff. Don't make the choice. --- ## Combining Altitudes Sometimes the best strategy is an idea from one altitude funding or enabling an idea from a higher one. A Monday Morning idea that generates data which proves the case for a Stretch idea. A This Quarter idea that builds the audience, coalition, or infrastructure a Moonshot idea would need. A pilot study at Altitude 2 that produces the preliminary results needed to justify the Altitude 4 grant proposal. A small policy experiment at Altitude 1 that generates the evidence base for the Altitude 3 legislative push. After presenting the four altitudes, look for these ladders: places where a lower-altitude idea is the first step toward a higher-altitude one. If you find one, note it. "The Monday Morning version of this (test it with 50 people manually) produces the data you'd need to justify the Stretch version (build it as a real system). One funds the other." These ladders are often more valuable than any single idea because they give the user a path from where they are to where they want to be with a ramp instead of a cliff. --- ## When to Use This Skill **Use it when:** - The user needs to make a decision about how ambitious to be - Every idea in a brainstorm is at the same risk level and there's nothing to compare against - The user says "give me the safe version and the bold version" - A team, lab, or committee is split between conservative and aggressive approaches and needs to see the full spectrum to find common ground - The user is stuck between "this isn't big enough to matter" and "this is too big to actually do" **Don't use it when:** - The user already knows their altitude and just needs ideas at that level - The task is purely creative generation with no implementation dimension (naming, taglines, hooks — use Short Think) - Speed matters more than range (use Short Think) --- ## Relationship to Other Skills **After the Guilford Engine:** The Guilford Engine produces ideas with feasibility tags (Now / Soon / Stretch / Moonshot). Wild to Mild goes deeper — it doesn't just tag ideas, it deliberately generates at each altitude separately, producing ideas that are native to that risk level rather than tagged after the fact. **After Think Wrong:** Think Wrong tends to produce ideas at Altitude 3-4 (Stretch and Moonshot). Wild to Mild can backfill the Monday Morning and This Quarter versions that make the bold idea actionable. "Think Wrong found the destination. Wild to Mild builds the road." **After Blind Spot Scan:** If the Blind Spot Scan reveals that all ideas are at one altitude (all safe, or all moonshots), Wild to Mild fills the missing altitudes. **With Strip Down:** Extract the desire. Then generate at all four altitudes from the same desire. The desire stays constant. The ambition level changes. This shows the user what their desire looks like as a Monday task versus a multi-year transformation. Wild to Mild is the only skill in the stack that explicitly addresses the strategic question underneath every brainstorm: "Not just what should we do, but how much should we bet?"
GitHub에서 보기