Top 3

Quick facts

Mechanism: Produces a worked three-framework analysis of a topic.

Produces: applied-framework-set

Evidence: C · Domain: Meta Thinking and Reflection

Use when a decision, problem, or stuck point deserves more than one lens and the goal is worked output now rather than a plan of which methods to run later. Use when several frameworks plausibly fit and the most relevant few should be ranked, executed, and integrated into a single read.

Most thinking tools name or recommend a method. This one runs three. It ranks the library’s frameworks for relevance to your topic, takes the most relevant three, applies each to the topic so it emits that framework’s real artifact, and then reconciles the three into one integrated read. The distinct move is execution at a fixed count of three, plus the cross-framework synthesis: not “here are the methods to consider”, but “here is what three fitting methods, worked end to end, jointly say about this”.

When to Use

• A real decision, problem, or stuck point where one lens is not enough and you want worked output now, not a routing plan.
• Several frameworks plausibly apply and you want the most relevant few ranked, executed, and integrated rather than chosen for you.
• You want the combined read: where the strongest-fitting methods agree, conflict, and converge on a next move.

When NOT to Use

• You do not yet know whether you need three frameworks, or might need none. Use think-framework-advisor; it subtracts to the fewest fitting moves (often zero to two). This skill force-applies exactly three and will manufacture work on a problem that needs one move or none.
• You already know the single framework you want. Run that skill directly, for example think-premortem. This skill adds two frameworks you did not ask for.
• A known multi-step chain already fits the job. Use the relevant recipe (reframe-problem, expand-options, stress-test-decision, audit-reasoning); recipes are curated, sequence-checked chains. This skill is an ad-hoc relevance-ranked set with no validated sequence.
• You are stuck and want unexpected lenses to break a frozen framing. That is the opposite selection rule; use think-random-frameworks.
• Not a thinking task (lookup, drafting, coding): redirect.

Instructions

When asked for the top three frameworks applied to a topic, follow these steps:

1. Parse the topic. Restate the situation in one or two sentences. If the input is under about 15 words or carries no concrete signal, ask one clarifying question, then proceed.
2. Run the shared engine in RANK mode. Follow references/engine.md: read the corpus, score every entry for relevance, take the top three (deduped by cognitive job), and apply each framework to the topic so it emits that framework’s real artifact, honoring each one’s When NOT to Use.
3. Synthesize. Reconcile the three artifacts into one integrated read: where they converge, where they conflict, and the single most load-bearing conclusion for the topic.

Output Format

Use the template in references/TEMPLATE.md. The deliverable is the three named-and-justified frameworks, the three filled artifacts, and the cross-framework synthesis, not a prose essay and not a list of recommendations.

Quality Checklist

Before finalizing, verify:

• Exactly three frameworks, each with a name that exists in the recommendable corpus and a one-line reason it ranked.
• Each framework was applied, emitting its real artifact, not merely named or recommended.
• The three do different cognitive jobs (no near-duplicate lenses); any when-NOT no-fit was swapped out and noted.
• A cross-framework synthesis reconciles convergence, conflict, and the load-bearing conclusion.
• No tier inflation: each framework’s evidence tier is carried honestly; applying three does not multiply confidence.

Evidence

Tier C (conceptually plausible, under-tested). Applying a fitting structured method to a decision is well-supported per method, but neither the relevance ranking nor the fixed-three-then-synthesize contract has been measured, in humans or in AI use. Treat the selection as a useful starting hypothesis and trust the individual frameworks’ own tiers (carried from each skill) over this meta-skill’s. Full grading and caveats: evidence/dossier.md.

Examples

See references/EXAMPLE.md for a completed run that ranks, applies, and synthesizes three frameworks on a real topic.

Deep dive: worked example

A full worked run (the shared Northwind scenario)

Example run: think-top3

A compressed worked run. The real artifacts would be fully filled from each framework’s own template; they are abbreviated here to show the shape.

Topic

“We are deciding whether to sunset our free tier next quarter. It is hard to reverse once existing free users are migrated or churned, the team is split, and a wrong call costs us months and goodwill.”

Selected frameworks (ranked)

#	Framework	Why it ranked	Tier
1	think-one-way-vs-two-way-door	The topic flags irreversibility (“hard to reverse once migrated”); the dominant job is calibrating how much rigor the decision deserves	P
2	think-premortem	A stakes-bearing, contested call with a real downside; surfacing failure modes before committing is a direct fit	S/M
3	think-decision-option-review	The team is split across options (sunset, keep, restructure); a structured option comparison does a distinct job from the other two	P

Recipe note: the stress-test-decision recipe also fits a chosen option; here no option is chosen yet, so the three individual frameworks are used instead.

Applied artifacts (abbreviated)

1. think-one-way-vs-two-way-door

Classification: one-way door (migrating or churning free users is not cleanly reversible). Implication: this warrants the heavier, slower decision process, not a quick call. Pre-commit test named: a reversible pilot (sunset for new signups only) would convert it toward a two-way door.

2. think-premortem

Imagined failure (“it is 6 months later and this was a disaster”): top failure modes - free-to-paid conversion assumption was wrong; goodwill and word-of-mouth loss exceeded modeled churn; a competitor’s free tier absorbed the leavers. Leading indicators and a kill-criterion attached to each.

3. think-decision-option-review

Options scored against weighted criteria (revenue lift, churn risk, reversibility, team capacity): keep-as-is, full-sunset, restructure-to-trial. Restructure-to-trial scores highest on reversibility while capturing most of the revenue case.

Cross-framework synthesis

• Converge: all three point away from a clean full-sunset now: door analysis says go slow, premortem says the conversion and goodwill assumptions are the fragile load-bearing beliefs, and the option review favors the reversible restructure.
• Conflict: the option review’s revenue case pulls toward acting soon; the door and premortem pull toward a reversible pilot first.
• Load-bearing conclusion: the decision hinges on one untested belief - that free users convert rather than churn - and there is a reversible way to test it before the one-way move.
• Next move: run a time-boxed trial-tier pilot for new signups, instrument conversion vs churn, and revisit the full sunset with real data.

Tiers carried honestly: P, S/M, P. The ranking and the fixed-three contract are tier C; the per-framework conclusions carry the frameworks’ own tiers.

Grounding: the full evidence dossier

What the research does and does not show, with graded sources

Evidence dossier: think-top3

What it is

A meta-skill that ranks the library’s thinking frameworks for relevance to a topic, applies the top three so each emits its real artifact, and reconciles the three into one cross-framework synthesis. It is an applicator over the shipped corpus, not a thinking method in its own right.

The mechanism (the durable move)

Two moves that the individual skills and the advisor do not combine: (1) relevance ranking the catalog to a fixed N of three, and (2) executing all three and synthesizing across them. The advisor recommends and subtracts; the individual skills each run one method. The distinct contribution is “three fitting methods, worked end to end, integrated into a single read”.

What the evidence says (honest grade: C)

The per-method claim - that applying a fitting structured method to a decision helps - is the evidence each underlying skill carries (tiers S through P), not this skill’s. This skill adds two unmeasured contracts on top:

• the relevance ranking of methods to a topic, and
• the fixed-three-then-synthesize protocol. Neither has been measured, in humans or in AI use. There is no study showing that exactly three is better than two or four, or that an automatic ranking beats the user’s own judgement of fit. The cross-synthesis is plausibly valuable (triangulating across lenses is a recognized practice) but untested in this packaged form.

Honest tier: C (conceptually plausible, under-tested). Transferred evidence: the value rides on the applied frameworks’ own grades, which are carried through and never inflated. This meta-skill’s selection and N=3 contract carry no validation.

When it misleads / when NOT to use

• On a problem that needs one move or none: it manufactures work (the advisor is the right tool, it subtracts).
• When the user already knows the method: it adds two unrequested frameworks.
• When a curated, sequence-checked recipe fits: the recipe has a validated order; this skill claims none.
• Treating “three frameworks agreed” as three independent confirmations: it is not. Over-claim is the failure mode; tiers are carried, never multiplied.

Lineage and who to read

Triangulation and multi-method analysis (using several independent lenses and reconciling them) is the intellectual root. The honest framing here is deliberately modest: the value is in the applied artifacts (each carrying its own evidence) and the synthesis, not in any claim that ranking-plus-three is itself a validated procedure.