Cognitive Mode-Switch Words
Words that don't describe what you want — they rewire how you approach the problem.
Quick Reference: Word → Behavior
When the user uses these words, activate the corresponding thinking mode:
Depth & Rigor
| Word |
Thinking Mode |
Invoke Skill |
| Rigorous |
Proof-like thinking — every step must earn its place |
rigorous-thinking |
| First-principles |
Derive from ground truth, strip borrowed reasoning |
rigorous-thinking |
| Mechanistic |
Demand causal chains: "A causes B because..." |
rigorous-thinking |
| Axiomatic |
Each claim follows logically from stated premises |
rigorous-thinking |
| Falsifiable |
Flip to "what would prove this wrong?" |
rigorous-thinking |
| Steelman |
Construct strongest possible version of opposing view |
rigorous-thinking, rigorous-exchange |
| Non-trivial |
Skip basics, assume expertise, go deeper |
— |
| Exhaustive |
Complete coverage, no blind spots, breadth before depth |
— |
| Necessary and sufficient |
Exact boundaries — no more, no less |
claim-validator |
| Load-bearing |
Identify the single assumption everything depends on |
rigorous-thinking |
| Dispositive |
Find the one fact that settles the debate |
rigorous-thinking |
| Ultra think hard |
Maximum cognitive effort, no shortcuts anywhere |
rigorous-thinking |
Creative Reframing
| Word |
Thinking Mode |
Invoke Skill |
| Reimagine |
Break fixed mental models — clean-slate vision |
ideacritic |
| Orthogonal |
Think perpendicular to the current frame |
ideacritic |
| Contrarian |
Deliberate inversion of the popular view |
ideacritic |
| Subversive |
Question the premise itself, not just disagree |
ideacritic |
| Invert |
Instead of "how to succeed," ask "how to guarantee failure" |
inversion-analysis |
| Counterfactual |
"What would have happened if X didn't occur?" |
inversion-analysis |
| Zero-based |
Start from scratch — if you had none of this, what would you build? |
ideacritic |
| Provocative |
Bold, non-consensus ideas you'd never say in a meeting |
ideacritic |
| Transmute |
Convert a weakness into a strength |
ideacritic |
| Reframe |
Re-examine whether you're solving the right problem |
ideacritic |
| Degenerate case |
Push variables to 0, 1, or infinity to reveal hidden structure |
— |
| Adjacent possible |
Constrain creativity to what's buildable from current position |
— |
Precision & Mathematical
| Word |
Thinking Mode |
Invoke Skill |
| Probabilities/Distribution |
Full statistical reasoning — likelihoods, not certainties |
claim-validator |
| Bounded |
Force upper and lower limits on any assertion |
claim-validator, unit-economics-validator |
| Monotonic |
Does this always increase/decrease, or are there reversals? |
— |
| Asymptotic |
What happens as scale approaches infinity? |
— |
| Invariant |
Identify the constant in a sea of variables |
— |
| Order of magnitude |
Calibration without false precision |
unit-economics-validator |
| Sensitivity analysis |
Which variable, if wrong by 2x, breaks the conclusion? |
unit-economics-validator |
| Convex |
Is the payoff shape concave (diminishing) or convex (accelerating)? |
— |
| Marginal |
What does one more unit buy you? Not average, incremental. |
pricing-strategy |
| Dimensionality |
How many independent axes actually matter? |
— |
| Closed-form |
Demand a computable formula, not hand-waving |
— |
Tradeoff & Decision
| Word |
Thinking Mode |
Invoke Skill |
| Sacrifice |
Name what you're willing to lose — make tradeoffs explicit |
operator-playbook |
| Irreversible |
Can we undo this or not? One-way vs two-way doors |
operator-playbook |
| Binding constraint |
Identify the single bottleneck |
systems-decompose |
| Pareto |
What 20% of effort gets 80% of the result? |
operator-playbook |
| Second-order effects |
What happens after the thing happens? Ripple effects. |
systems-review |
| Opportunity cost |
Is this better than the next best alternative? |
operator-playbook |
| Regret minimization |
Which choice do you regret least looking back from 10 years out? |
operator-playbook |
| Non-negotiable |
What will you absolutely not bend on? |
— |
| Asymmetric |
Where upside and downside are mismatched |
— |
| Reversibility gradient |
Not binary — place decision on the spectrum |
— |
Quality & Craft
| Word |
Thinking Mode |
Invoke Skill |
| Elegant |
Simplicity, completeness, and surprise in one solution |
code-review, software-architecture |
| Idiomatic |
Domain-native solutions, not generic ones |
code-review |
| Canonical |
The accepted best practice, not improvisation |
code-review |
| Production-grade |
Error handling, edge cases, monitoring, logging |
code-review, security-review |
| Battle-tested |
Proven approaches over clever theory |
software-architecture |
| Minimalist |
Remove everything that isn't essential |
simplify |
| Composable |
Modular design where pieces can be mixed and matched |
software-architecture |
| Tasteful |
Aesthetic reasoning — what feels right to an expert? |
— |
| Crisp |
Precision in language — every word must earn its place |
deslop |
Systems Thinking
| Word |
Thinking Mode |
Invoke Skill |
| Emergent |
What arises from interactions that no individual part possesses? |
systems-decompose |
| Cascading |
Maps domino effects through a system |
systems-review |
| Equilibrium |
Is this equilibrium stable or unstable? What disrupts it? |
— |
| Phase transition |
At what threshold does the system fundamentally shift? |
— |
| Feedback loop |
Is this positive (amplifying) or negative (stabilizing)? |
systems-decompose |
| Attractor |
Where does this naturally converge? |
— |
| Entropy |
What's degrading over time? What maintenance prevents collapse? |
techdebt |
Epistemic Calibration
| Word |
Thinking Mode |
Invoke Skill |
| Bayesian |
Explicit priors, evidence weighting, posterior updates |
rigorous-thinking |
| Calibrated |
Honest probability assignment to claims |
rigorous-thinking |
| Crux |
Single disagreement that, if settled, resolves everything |
rigorous-exchange |
| Pre-mortem |
Assume this failed. Now explain why. |
inversion-analysis |
| Evidence-weighted |
Which signals carry the most weight and why? |
rigorous-thinking |
| Adversarial |
Assume someone is actively trying to break this |
security-review |
| Sparse |
Strip away everything except what carries information |
rigorous-thinking |
| Steel-thread |
Thinnest possible end-to-end proof of concept |
— |
| Epistemic humility |
Honest "here's what I don't know" disclosure |
rigorous-thinking |
| Signal-to-noise |
What fraction of this information changes decisions? |
rigorous-thinking |
Power Combos
The highest-leverage technique: pair a depth word with a constraint word.
| Combo |
Effect |
Invoke |
| Rigorous + Elegant |
Airtight reasoning that's also beautiful and simple |
rigorous-thinking + code-review |
| Reimagine + First-principles |
Clean-slate vision built from ground truth |
ideacritic + rigorous-thinking |
| Ultra think hard + Sacrifice |
Maximum depth applied to what must be cut |
rigorous-thinking + operator-playbook |
| Falsifiable + Bounded |
Testable claims with defined ranges |
rigorous-thinking + claim-validator |
| Adversarial + Exhaustive |
Complete threat surface mapping |
security-review |
| Contrarian + Evidence-weighted |
Bold takes backed by real data |
ideacritic + rigorous-thinking |
| Invert + Pre-mortem |
Failure-first scenario planning |
inversion-analysis |
| Sparse + Dispositive |
Cut to the one thing that decides everything |
rigorous-thinking |
| Distribution + Sensitivity analysis |
Full outcome range with fragility mapping |
unit-economics-validator |
Usage
This skill auto-invokes when the user uses trigger words. It then:
- Activates the thinking mode corresponding to the word(s)
- Invokes relevant skills if deeper analysis is needed
- Applies the cognitive constraint to all reasoning in the response
Examples
User: "Be rigorous: why won't this architecture scale?"
→ Invoke rigorous-thinking, apply proof-like reasoning
User: "Give me the contrarian take on our pricing."
→ Invoke ideacritic, deliberately invert the popular view
User: "Ultra think hard + sacrifice: what do we cut?"
→ Maximum cognitive effort applied to naming tradeoffs
User: "Sensitivity analysis on our unit economics."
→ Invoke unit-economics-validator, identify fragile assumptions
The Meta-Pattern
Words that constrain the solution space in a non-obvious way always outperform words that merely describe desired quality.
- ❌ "Good" is useless
- ✅ "Elegant" rewires the entire approach
- ✅ "Falsifiable" flips from prove-right to prove-wrong
- ✅ "Sparse" strips away noise before answering
When you hear a trigger word, change HOW you think, not just WHAT you output.
1---2name: ai-lingo3description: Meta-skill that maps cognitive mode-switch words to appropriate thinking behaviors and skills. **AUTO-INVOKE when user uses these trigger patterns:** - Depth: "rigorous", "first-principles", "mechanistic", "axiomatic", "falsifiable", "steelman", "non-trivial", "exhaustive", "load-bearing", "dispositive", "ultra think hard" - Reframe: "reimagine", "orthogonal", "contrarian", "subversive", "invert", "counterfactual", "zero-based", "provocative", "transmute", "reframe" - Precision: "probabilities", "bounded", "monotonic", "asymptotic", "invariant", "order of magnitude", "sensitivity analysis", "convex", "distribution", "marginal", "dimensionality" - Tradeoff: "sacrifice", "irreversible", "binding constraint", "pareto", "second-order", "opportunity cost", "regret minimization", "non-negotiable", "asymmetric" - Quality: "elegant", "idiomatic", "canonical", "production-grade", "battle-tested", "minimalist", "composable", "tasteful", "crisp" - Systems: "emergent", "cascading", "equilibrium", "phase transition", "fee4---56# Cognitive Mode-Switch Words78Words that don't describe what you want — they rewire **how** you approach the problem.910## Quick Reference: Word → Behavior1112When the user uses these words, activate the corresponding thinking mode:1314### Depth & Rigor1516| Word | Thinking Mode | Invoke Skill |17|------|---------------|--------------|18| **Rigorous** | Proof-like thinking — every step must earn its place | `rigorous-thinking` |19| **First-principles** | Derive from ground truth, strip borrowed reasoning | `rigorous-thinking` |20| **Mechanistic** | Demand causal chains: "A causes B because..." | `rigorous-thinking` |21| **Axiomatic** | Each claim follows logically from stated premises | `rigorous-thinking` |22| **Falsifiable** | Flip to "what would prove this wrong?" | `rigorous-thinking` |23| **Steelman** | Construct strongest possible version of opposing view | `rigorous-thinking`, `rigorous-exchange` |24| **Non-trivial** | Skip basics, assume expertise, go deeper | — |25| **Exhaustive** | Complete coverage, no blind spots, breadth before depth | — |26| **Necessary and sufficient** | Exact boundaries — no more, no less | `claim-validator` |27| **Load-bearing** | Identify the single assumption everything depends on | `rigorous-thinking` |28| **Dispositive** | Find the one fact that settles the debate | `rigorous-thinking` |29| **Ultra think hard** | Maximum cognitive effort, no shortcuts anywhere | `rigorous-thinking` |3031### Creative Reframing3233| Word | Thinking Mode | Invoke Skill |34|------|---------------|--------------|35| **Reimagine** | Break fixed mental models — clean-slate vision | `ideacritic` |36| **Orthogonal** | Think perpendicular to the current frame | `ideacritic` |37| **Contrarian** | Deliberate inversion of the popular view | `ideacritic` |38| **Subversive** | Question the premise itself, not just disagree | `ideacritic` |39| **Invert** | Instead of "how to succeed," ask "how to guarantee failure" | `inversion-analysis` |40| **Counterfactual** | "What would have happened if X didn't occur?" | `inversion-analysis` |41| **Zero-based** | Start from scratch — if you had none of this, what would you build? | `ideacritic` |42| **Provocative** | Bold, non-consensus ideas you'd never say in a meeting | `ideacritic` |43| **Transmute** | Convert a weakness into a strength | `ideacritic` |44| **Reframe** | Re-examine whether you're solving the right problem | `ideacritic` |45| **Degenerate case** | Push variables to 0, 1, or infinity to reveal hidden structure | — |46| **Adjacent possible** | Constrain creativity to what's buildable from current position | — |4748### Precision & Mathematical4950| Word | Thinking Mode | Invoke Skill |51|------|---------------|--------------|52| **Probabilities/Distribution** | Full statistical reasoning — likelihoods, not certainties | `claim-validator` |53| **Bounded** | Force upper and lower limits on any assertion | `claim-validator`, `unit-economics-validator` |54| **Monotonic** | Does this always increase/decrease, or are there reversals? | — |55| **Asymptotic** | What happens as scale approaches infinity? | — |56| **Invariant** | Identify the constant in a sea of variables | — |57| **Order of magnitude** | Calibration without false precision | `unit-economics-validator` |58| **Sensitivity analysis** | Which variable, if wrong by 2x, breaks the conclusion? | `unit-economics-validator` |59| **Convex** | Is the payoff shape concave (diminishing) or convex (accelerating)? | — |60| **Marginal** | What does one more unit buy you? Not average, incremental. | `pricing-strategy` |61| **Dimensionality** | How many independent axes actually matter? | — |62| **Closed-form** | Demand a computable formula, not hand-waving | — |6364### Tradeoff & Decision6566| Word | Thinking Mode | Invoke Skill |67|------|---------------|--------------|68| **Sacrifice** | Name what you're willing to lose — make tradeoffs explicit | `operator-playbook` |69| **Irreversible** | Can we undo this or not? One-way vs two-way doors | `operator-playbook` |70| **Binding constraint** | Identify the single bottleneck | `systems-decompose` |71| **Pareto** | What 20% of effort gets 80% of the result? | `operator-playbook` |72| **Second-order effects** | What happens *after* the thing happens? Ripple effects. | `systems-review` |73| **Opportunity cost** | Is this better than the next best alternative? | `operator-playbook` |74| **Regret minimization** | Which choice do you regret least looking back from 10 years out? | `operator-playbook` |75| **Non-negotiable** | What will you absolutely not bend on? | — |76| **Asymmetric** | Where upside and downside are mismatched | — |77| **Reversibility gradient** | Not binary — place decision on the spectrum | — |7879### Quality & Craft8081| Word | Thinking Mode | Invoke Skill |82|------|---------------|--------------|83| **Elegant** | Simplicity, completeness, and surprise in one solution | `code-review`, `software-architecture` |84| **Idiomatic** | Domain-native solutions, not generic ones | `code-review` |85| **Canonical** | The accepted best practice, not improvisation | `code-review` |86| **Production-grade** | Error handling, edge cases, monitoring, logging | `code-review`, `security-review` |87| **Battle-tested** | Proven approaches over clever theory | `software-architecture` |88| **Minimalist** | Remove everything that isn't essential | `simplify` |89| **Composable** | Modular design where pieces can be mixed and matched | `software-architecture` |90| **Tasteful** | Aesthetic reasoning — what *feels* right to an expert? | — |91| **Crisp** | Precision in language — every word must earn its place | `deslop` |9293### Systems Thinking9495| Word | Thinking Mode | Invoke Skill |96|------|---------------|--------------|97| **Emergent** | What arises from interactions that no individual part possesses? | `systems-decompose` |98| **Cascading** | Maps domino effects through a system | `systems-review` |99| **Equilibrium** | Is this equilibrium stable or unstable? What disrupts it? | — |100| **Phase transition** | At what threshold does the system fundamentally shift? | — |101| **Feedback loop** | Is this positive (amplifying) or negative (stabilizing)? | `systems-decompose` |102| **Attractor** | Where does this naturally converge? | — |103| **Entropy** | What's degrading over time? What maintenance prevents collapse? | `techdebt` |104105### Epistemic Calibration106107| Word | Thinking Mode | Invoke Skill |108|------|---------------|--------------|109| **Bayesian** | Explicit priors, evidence weighting, posterior updates | `rigorous-thinking` |110| **Calibrated** | Honest probability assignment to claims | `rigorous-thinking` |111| **Crux** | Single disagreement that, if settled, resolves everything | `rigorous-exchange` |112| **Pre-mortem** | Assume this failed. Now explain why. | `inversion-analysis` |113| **Evidence-weighted** | Which signals carry the most weight and why? | `rigorous-thinking` |114| **Adversarial** | Assume someone is actively trying to break this | `security-review` |115| **Sparse** | Strip away everything except what carries information | `rigorous-thinking` |116| **Steel-thread** | Thinnest possible end-to-end proof of concept | — |117| **Epistemic humility** | Honest "here's what I don't know" disclosure | `rigorous-thinking` |118| **Signal-to-noise** | What fraction of this information changes decisions? | `rigorous-thinking` |119120---121122## Power Combos123124The highest-leverage technique: **pair a depth word with a constraint word.**125126| Combo | Effect | Invoke |127|-------|--------|--------|128| **Rigorous + Elegant** | Airtight reasoning that's also beautiful and simple | `rigorous-thinking` + `code-review` |129| **Reimagine + First-principles** | Clean-slate vision built from ground truth | `ideacritic` + `rigorous-thinking` |130| **Ultra think hard + Sacrifice** | Maximum depth applied to what must be cut | `rigorous-thinking` + `operator-playbook` |131| **Falsifiable + Bounded** | Testable claims with defined ranges | `rigorous-thinking` + `claim-validator` |132| **Adversarial + Exhaustive** | Complete threat surface mapping | `security-review` |133| **Contrarian + Evidence-weighted** | Bold takes backed by real data | `ideacritic` + `rigorous-thinking` |134| **Invert + Pre-mortem** | Failure-first scenario planning | `inversion-analysis` |135| **Sparse + Dispositive** | Cut to the one thing that decides everything | `rigorous-thinking` |136| **Distribution + Sensitivity analysis** | Full outcome range with fragility mapping | `unit-economics-validator` |137138---139140## Usage141142This skill auto-invokes when the user uses trigger words. It then:1431441. **Activates the thinking mode** corresponding to the word(s)1452. **Invokes relevant skills** if deeper analysis is needed1463. **Applies the cognitive constraint** to all reasoning in the response147148### Examples149150**User:** "Be rigorous: why won't this architecture scale?"151→ Invoke `rigorous-thinking`, apply proof-like reasoning152153**User:** "Give me the contrarian take on our pricing."154→ Invoke `ideacritic`, deliberately invert the popular view155156**User:** "Ultra think hard + sacrifice: what do we cut?"157→ Maximum cognitive effort applied to naming tradeoffs158159**User:** "Sensitivity analysis on our unit economics."160→ Invoke `unit-economics-validator`, identify fragile assumptions161162---163164## The Meta-Pattern165166Words that constrain the **solution space** in a non-obvious way always outperform words that merely describe desired **quality**.167168- ❌ "Good" is useless169- ✅ "Elegant" rewires the entire approach170- ✅ "Falsifiable" flips from prove-right to prove-wrong171- ✅ "Sparse" strips away noise before answering172173**When you hear a trigger word, change HOW you think, not just WHAT you output.**
sunnypatneedi