From reasoning-framework
Use when stress-testing a conclusion, auditing a complex claim, mapping a multi-actor system, or spotting motivated reasoning — when rigor matters more than speed. Skip for time-pressured decisions, mechanical tasks, lookups, or when the user wants a recommendation rather than a tour of considerations.
How this skill is triggered — by the user, by Claude, or both
Slash command
/reasoning-framework:reasoning-frameworkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill is for **analysis and validation** — stress-testing conclusions and mapping genuine complexity through a **pluralistic, multi-disciplinary** posture. It is not for discovery under time pressure, mechanical work, or decisions that need immediate action. It recruits a reasoning orientation that is already latent; it does not install one. If invoking it would slow a task without changin...
This skill is for analysis and validation — stress-testing conclusions and mapping genuine complexity through a pluralistic, multi-disciplinary posture. It is not for discovery under time pressure, mechanical work, or decisions that need immediate action. It recruits a reasoning orientation that is already latent; it does not install one. If invoking it would slow a task without changing the conclusion, don't.
Before applying anything below, answer plainly: am I about to produce the appearance of rigor or its substance? If the procedure would dress shallow analysis in deep-looking language — long lists, careful hedges, gestures at named "frameworks" — stop and answer the question directly. Naming a framework is not the same as using one. If the framework didn't change your conclusion, don't name it. This warning applies to every named source below.
Use dialectical reasoning, tetralemma analysis, Socratic questioning, and systems thinking as simultaneous recursive lenses with continuous cross-method feedback — insights from each method inform and reshape the others. Hold holistic awareness of contextual relationships: a claim is not just a proposition, it is embedded in time, system, and stakes. Contradiction is productive tension, not error to resolve. A meta-analysis layer attends to emergent patterns visible only across lenses; those cross-cutting patterns take priority over conclusions reached within any single lens. Empiricism anchors throughout as a continuous constraint, not a final filter.
The lenses are not a checklist. Activate what the subject requires; most questions need two or three.
Before engaging the content, surface the question's assumptions and framings. Consider what would make it dissolve rather than require answering. Identify whose interests the current framing serves and which actors it makes invisible. If the question rests on a false premise, name the premise and dissolve the question rather than answer it.
For a claim X, examine four positions:
Dwell in genuine contradictions rather than rushing to resolve them. Contradiction locates real complexity. Resolving too quickly is how complexity gets hidden.
Ask:
Watch especially for reification (treating an abstraction as a concrete entity) and composition / division (assuming what holds of the parts holds of the whole, or vice versa). Name the fallacy pattern when you see it in your own argument.
Follow second-order effects, feedback loops, missing actors, and leverage points:
Run the conclusion through multiple lenses and track what changes. This is the meta-analysis layer the orientation depends on.
Empiricism is not a filter applied at the end; it is a continuous constraint. Sagan's nine heuristics, compressed:
Additionally: flag motivated reasoning in others' arguments and your own; verify that claims survive contact with contrary evidence; test by consequences to those affected, not only by logical elegance.
The Buddha's instruction to the Kalamas: do not accept a claim merely because of —
Accept it after observation and analysis show it agrees with reason and is conducive to the good of those affected. The Kalama posture is not skeptical paralysis — it is the refusal to defer judgment about consequences to any source other than the world itself.
Distinguish, and label distinctly:
Not everything worth knowing survives propositional expression. Acknowledge that boundary instead of hiding it.
The product of this skill is not "a confident answer" by default. Often it is: what survives, what doesn't, and the question you should be asking instead.
Provides 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.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
npx claudepluginhub ravenoak/llm-skills --plugin reasoning-framework