{"id":"919c86f4-f586-4149-ad37-8193dceb9680","entityType":"agent","slug":"clawhub-deciqai-dual-system-thinking","name":"Dual-System Thinking (System 1 / System 2)","canonicalUrl":"https://www.xpersona.co/agent/clawhub-deciqai-dual-system-thinking","canonicalPath":"/agent/clawhub-deciqai-dual-system-thinking","generatedAt":"2026-10-11T08:42:06.100Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":null},"description":"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou... Skill: Dual-System Thinking (System 1 / System 2) Owner: deciqai Summary: Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:58:16.166Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/dual-system-thinking.json) v1.0","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:dual-system-thinking","sourceUrl":"https://clawhub.ai/deciqai/dual-system-thinking","homepage":"https://clawhub.ai/deciqai/skills/dual-system-thinking","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/deciqai/dual-system-thinking","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/deciqai/skills/dual-system-thinking","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou..."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":null},"stars":null,"forks":null,"downloads":1133,"packageName":null,"latestVersion":"1.0.5","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T06:29:45.690Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T06:29:45.704Z","lastCrawledAt":"2026-10-11T06:29:45.690Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T06:29:45.690Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.5","createdAt":"2026-07-16T17:58:16.166Z","changelog":"Description tail link + agents machine-readable metadata line (deciqai.com/s/dual-system-thinking.json)","fileCount":6,"zipByteSize":13204},{"version":"1.0.4","createdAt":"2026-07-10T10:25:16.360Z","changelog":"Add 2024-2026 AI-era worked example + updated sources","fileCount":6,"zipByteSize":12974},{"version":"1.0.3","createdAt":"2026-07-08T11:00:31.310Z","changelog":"Footer now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)","fileCount":5,"zipByteSize":9031},{"version":"1.0.2","createdAt":"2026-07-08T00:45:32.129Z","changelog":"Refreshed content + GitHub star link in footer","fileCount":5,"zipByteSize":8948},{"version":"1.0.1","createdAt":"2026-07-07T20:32:23.982Z","changelog":"Add catalog categories and topics","fileCount":5,"zipByteSize":8922},{"version":"1.0.0","createdAt":"2026-06-28T06:17:20.290Z","changelog":"Initial publish","fileCount":5,"zipByteSize":8972}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:dual-system-thinking","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T08:42:06.096Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":null},"readme":"Skill: Dual-System Thinking (System 1 / System 2)\n\nOwner: deciqai\n\nSummary: Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou...\n\nTags: latest:1.0.5\n\nVersion history:\n\nv1.0.5 | 2026-07-16T17:58:16.166Z | user\n\nDescription tail link + agents machine-readable metadata line (deciqai.com/s/dual-system-thinking.json)\n\nv1.0.4 | 2026-07-10T10:25:16.360Z | user\n\nAdd 2024-2026 AI-era worked example + updated sources\n\nv1.0.3 | 2026-07-08T11:00:31.310Z | user\n\nFooter now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)\n\nv1.0.2 | 2026-07-08T00:45:32.129Z | user\n\nRefreshed content + GitHub star link in footer\n\nv1.0.1 | 2026-07-07T20:32:23.982Z | user\n\nAdd catalog categories and topics\n\nv1.0.0 | 2026-06-28T06:17:20.290Z | user\n\nInitial publish\n\nArchive index:\n\nArchive v1.0.5: 6 files, 13204 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), examples/llms-as-system-1-ai-era-2024-2026.md (6894b), references/sources.md (1877b), skill-card.md (2733b), SKILL.md (7418b), _meta.json (139b)\n\nFile v1.0.5:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast. More: deciqai.com/c/dual-system-thinking\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n- Acting on a fast, fluent, confident AI/LLM answer (AI adoption, AI hype, \"the AI said so\") on a high-stakes call — when to force slow verification of a confident machine response\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n*→ 2026 lens: [LLMs as System 1 — when to force System 2 on a confident AI answer (2024–2026)](examples/llms-as-system-1-ai-era-2024-2026.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/dual-system-thinking** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\n*Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/dual-system-thinking.json*\n\nFile v1.0.5:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1784224696166\n}\n\nFile v1.0.5:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n- Ji, Z., Lee, N., Frieske, R., et al. (2023). \"Survey of Hallucination in Natural Language Generation.\" *ACM Computing Surveys*, 55(12), 1-38. LLMs as fluent-but-uncalibrated System-1-like responders (2024–2026 AI-era example).\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461, S.D.N.Y. (2023). Attorneys sanctioned for a filing containing AI-generated citations to non-existent cases — a widely reported high-stakes failure of un-verified, confident AI output.\n\nFile v1.0.5:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.5:examples/llms-as-system-1-ai-era-2024-2026.md\n\n# Method in Action: LLMs as System 1 — When to Force System 2 on a Confident AI Answer (2024–2026)\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nBy 2024–2026, hundreds of millions of people were routing everyday judgments through large language models (ChatGPT, Claude, Gemini, and copilots embedded in search, IDEs, and office software). A useful and increasingly common framing among researchers and practitioners: a modern LLM behaves like an externalized **System 1** — it returns a fast, fluent, confident answer by pattern-completion, with no felt effort and no built-in signal for when it is guessing. The fluency is the trap. A wrong answer and a right answer arrive in the same authoritative prose, at the same speed, with the same surface confidence.\n\nThis example runs that situation through the skill's own Process. The \"decision\" is not \"what did the AI say\" but *\"should I act on this AI answer as-is, or recruit System 2 to verify it?\"*\n\n## The Analogy (and its limits)\n\nAn LLM maps onto **System 1** in the ways that matter operationally: automatic, effortless from the user's side, pattern-driven, fast, and frequently overconfident. It is *not* literally a human System 1, and it can be *made* to do System-2-like work — chain-of-thought prompting, tool use, retrieval, self-critique, and the \"reasoning\" model modes released across 2024–2025 all push the model toward slower, more deliberate, checkable output. The lesson is not \"AI = System 1, full stop.\" It is: **the default, un-recruited mode of an LLM is System-1-shaped, and the human on the other end is the System 2 that has to decide when to recruit deliberation — the model's or their own.**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer.** Decision: *do I ship / send / act on this LLM output?* The \"initial answer\" is the model's response, which arrives in seconds, reads as confident, and produces almost no felt effort in the user. Note the tell: your own confidence tends to be borrowed from the *fluency* of the text, not from any verification you performed.\n\n**Step 2 — Classify system.** The output is quintessential System 1: quick, automatic, confident, effortless. Critically, this holds even when the content is a fabrication. LLMs produce **hallucinations** (also called confabulations) — fluent, plausible, well-formatted statements that are simply false: invented citations, non-existent case law, wrong API signatures, made-up statistics. Fluency is not calibration. Treat the response as a System 1 output regardless of how authoritative it sounds.\n\n**Step 3 — Recruit decision.** Apply the skill's stakes × familiarity matrix — but read \"familiarity\" as *the model's* reliability in this exact domain *and* your own ability to check it:\n\n| Stakes | Domain the AI is reliable in? | Recruit System 2? |\n|---|---|---|\n| Low (draft an email, brainstorm) | High | No — let System 1 run |\n| Low | Low | Optional |\n| High (legal filing, medical, financial, production code, published fact) | High | Yes — verify |\n| High | Low (novel, niche, or high-hallucination-risk) | Mandatory — verify every load-bearing claim |\n\nThe documented failure modes cluster in the bottom-right cell: a lawyer files a brief citing cases the model invented; a developer ships an API call the model confabulated; a student submits a fabricated statistic. In several widely reported 2023–2024 U.S. court incidents, attorneys were sanctioned for filings containing AI-generated citations to cases that did not exist — a textbook high-stakes, low-verification System 1 error.\n\n**Step 4 — Recruit via procedure.** Concretely, for an LLM answer this means:\n- **Force the model's own System 2:** ask it to show its reasoning, to cite sources, to state its confidence, and to argue the *counter*-case (\"what would make this answer wrong?\"). Reasoning-mode / extended-thinking settings are System-2 recruitment at the tool level.\n- **Ground it:** require retrieval or a link to a primary source, then check the source actually says what the model claims. Do not accept a citation you have not opened.\n- **Verify externally:** run the code, check the number against the primary document, get a second model or a human expert to review.\n- **Premortem:** \"if I act on this and it's wrong, what breaks, and can I reverse it?\" Irreversible + high-stakes → verify before acting, always.\n\n**Step 5 — Compare outputs.** Compare the model's fast answer (S1) against your deliberate check (S2). If verification agrees → high confidence, act. If they diverge → S2 wins for any high-stakes or novel claim; the fluent original was a pattern-match, not a proof. Document *why* it diverged (hallucinated citation? outdated training data? plausible-but-wrong pattern?). The dangerous case is silent agreement-by-default: accepting the answer *because* it sounded right, which is exactly Kahneman's \"the way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n\n**Step 6 — Calibrate.** Keep a running log by domain: where does this model, in your hands, reliably get it right, and where has it burned you? High-calibration domains (e.g., boilerplate code, summarizing text you supply, ideation) earn more System-1 trust. Low-calibration domains (exact citations, current events past the training cutoff, precise numbers, niche law/medicine) get mandatory System 2 every time. This mirrors the human rule — trained System 1 is valuable *inside* its calibrated domain and dangerous outside it — now applied to a tool whose calibrated domain you have to learn empirically.\n\n## The Core Lesson\n\nThe AI era does not retire dual-system thinking — it makes it operationally urgent at population scale. The LLM is a spectacularly capable System 1 that you can rent by the token. Its greatest risk is not that it is often wrong, but that it is **confidently, fluently wrong in the same voice it uses when it is right**, stripping away the natural hesitation cues (a human hedging, pausing, saying \"I'm not sure\") that normally prompt us to slow down. The human's irreducible job is to be the System 2: to decide, per the stakes × reliability matrix, when to accept the fast answer and when to force verification — the model's or your own.\n\n*Sources: Kahneman, D. (2011), *Thinking, Fast and Slow* (System 1 / System 2 synthesis; \"feeling right\" vs \"being right\"). Ji, Z. et al. (2023), \"Survey of Hallucination in Natural Language Generation,\" *ACM Computing Surveys* 55(12) — defines and surveys LLM hallucination/confabulation. Widely reported 2023–2024 U.S. legal sanctions over AI-fabricated case citations (e.g., *Mata v. Avianca*, S.D.N.Y., 2023). OpenAI and Anthropic public documentation (2024–2025) on reasoning/extended-thinking modes and tool use as means of eliciting more deliberate, checkable model output.*\n\nFile v1.0.5:skill-card.md\n\n## Description:\n\nThis skill helps an agent slow down consequential or unfamiliar decisions by identifying fast System 1 intuitions, recruiting deliberate System 2 checks, and documenting the final rationale.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[deciqai](https://clawhub.ai/user/deciqai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nEmployees, external users, and AI agents use this skill to stress-test high-stakes decisions that feel obvious, especially under time pressure, fatigue, team convergence, unfamiliar domains, or reliance on fluent AI answers.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Decision logs can contain confidential business, personal, legal, medical, financial, or employment information.\n\nMitigation: Keep logs out of shared skill files and public notes unless they have been intentionally sanitized.\n\nRisk: The skill may be used around high-stakes decisions where coaching output is not a substitute for qualified review.\n\nMitigation: Use the worksheet as a decision-review aid and verify load-bearing claims with primary sources or appropriate domain experts.\n\nRisk: Fluent AI answers can be accepted too quickly when they appear confident.\n\nMitigation: Apply the skill's System 2 recruitment step before acting on high-stakes or unfamiliar AI-generated claims.\n\n## Reference(s):\n\n- [Primary sources for dual-system-thinking](references/sources.md)\n- [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)\n- [LLMs as System 1: When to Force System 2 on a Confident AI Answer](examples/llms-as-system-1-ai-era-2024-2026.md)\n- [Dual-System Thinking skill page](https://clawhub.ai/deciqai/skills/dual-system-thinking)\n- [deciqAI skill runtime page](https://www.deciqai.com/c/dual-system-thinking)\n- [Machine-readable skill metadata](https://www.deciqai.com/s/dual-system-thinking.json)\n- [deciqAI Knowledge Skills repository cited by artifact](https://github.com/deciqAI/knowledge-skills)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, guidance]\n\n**Output Format:** [Markdown decision-review worksheet with prompts, comparison fields, and a calibration-log template.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Coach-mode responses may pause at explicit WAIT steps until the user provides decision details.]\n\n## Skill Version(s):\n\n1.0.5 (source: server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.4: 6 files, 12974 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), examples/llms-as-system-1-ai-era-2024-2026.md (6894b), references/sources.md (1877b), skill-card.md (2376b), SKILL.md (7267b), _meta.json (139b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast.\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n- Acting on a fast, fluent, confident AI/LLM answer (AI adoption, AI hype, \"the AI said so\") on a high-stakes call — when to force slow verification of a confident machine response\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n*→ 2026 lens: [LLMs as System 1 — when to force System 2 on a confident AI answer (2024–2026)](examples/llms-as-system-1-ai-era-2024-2026.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 223 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/dual-system-thinking** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.4:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1783679116360\n}\n\nFile v1.0.4:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n- Ji, Z., Lee, N., Frieske, R., et al. (2023). \"Survey of Hallucination in Natural Language Generation.\" *ACM Computing Surveys*, 55(12), 1-38. LLMs as fluent-but-uncalibrated System-1-like responders (2024–2026 AI-era example).\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461, S.D.N.Y. (2023). Attorneys sanctioned for a filing containing AI-generated citations to non-existent cases — a widely reported high-stakes failure of un-verified, confident AI output.\n\nFile v1.0.4:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.4:examples/llms-as-system-1-ai-era-2024-2026.md\n\n# Method in Action: LLMs as System 1 — When to Force System 2 on a Confident AI Answer (2024–2026)\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nBy 2024–2026, hundreds of millions of people were routing everyday judgments through large language models (ChatGPT, Claude, Gemini, and copilots embedded in search, IDEs, and office software). A useful and increasingly common framing among researchers and practitioners: a modern LLM behaves like an externalized **System 1** — it returns a fast, fluent, confident answer by pattern-completion, with no felt effort and no built-in signal for when it is guessing. The fluency is the trap. A wrong answer and a right answer arrive in the same authoritative prose, at the same speed, with the same surface confidence.\n\nThis example runs that situation through the skill's own Process. The \"decision\" is not \"what did the AI say\" but *\"should I act on this AI answer as-is, or recruit System 2 to verify it?\"*\n\n## The Analogy (and its limits)\n\nAn LLM maps onto **System 1** in the ways that matter operationally: automatic, effortless from the user's side, pattern-driven, fast, and frequently overconfident. It is *not* literally a human System 1, and it can be *made* to do System-2-like work — chain-of-thought prompting, tool use, retrieval, self-critique, and the \"reasoning\" model modes released across 2024–2025 all push the model toward slower, more deliberate, checkable output. The lesson is not \"AI = System 1, full stop.\" It is: **the default, un-recruited mode of an LLM is System-1-shaped, and the human on the other end is the System 2 that has to decide when to recruit deliberation — the model's or their own.**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer.** Decision: *do I ship / send / act on this LLM output?* The \"initial answer\" is the model's response, which arrives in seconds, reads as confident, and produces almost no felt effort in the user. Note the tell: your own confidence tends to be borrowed from the *fluency* of the text, not from any verification you performed.\n\n**Step 2 — Classify system.** The output is quintessential System 1: quick, automatic, confident, effortless. Critically, this holds even when the content is a fabrication. LLMs produce **hallucinations** (also called confabulations) — fluent, plausible, well-formatted statements that are simply false: invented citations, non-existent case law, wrong API signatures, made-up statistics. Fluency is not calibration. Treat the response as a System 1 output regardless of how authoritative it sounds.\n\n**Step 3 — Recruit decision.** Apply the skill's stakes × familiarity matrix — but read \"familiarity\" as *the model's* reliability in this exact domain *and* your own ability to check it:\n\n| Stakes | Domain the AI is reliable in? | Recruit System 2? |\n|---|---|---|\n| Low (draft an email, brainstorm) | High | No — let System 1 run |\n| Low | Low | Optional |\n| High (legal filing, medical, financial, production code, published fact) | High | Yes — verify |\n| High | Low (novel, niche, or high-hallucination-risk) | Mandatory — verify every load-bearing claim |\n\nThe documented failure modes cluster in the bottom-right cell: a lawyer files a brief citing cases the model invented; a developer ships an API call the model confabulated; a student submits a fabricated statistic. In several widely reported 2023–2024 U.S. court incidents, attorneys were sanctioned for filings containing AI-generated citations to cases that did not exist — a textbook high-stakes, low-verification System 1 error.\n\n**Step 4 — Recruit via procedure.** Concretely, for an LLM answer this means:\n- **Force the model's own System 2:** ask it to show its reasoning, to cite sources, to state its confidence, and to argue the *counter*-case (\"what would make this answer wrong?\"). Reasoning-mode / extended-thinking settings are System-2 recruitment at the tool level.\n- **Ground it:** require retrieval or a link to a primary source, then check the source actually says what the model claims. Do not accept a citation you have not opened.\n- **Verify externally:** run the code, check the number against the primary document, get a second model or a human expert to review.\n- **Premortem:** \"if I act on this and it's wrong, what breaks, and can I reverse it?\" Irreversible + high-stakes → verify before acting, always.\n\n**Step 5 — Compare outputs.** Compare the model's fast answer (S1) against your deliberate check (S2). If verification agrees → high confidence, act. If they diverge → S2 wins for any high-stakes or novel claim; the fluent original was a pattern-match, not a proof. Document *why* it diverged (hallucinated citation? outdated training data? plausible-but-wrong pattern?). The dangerous case is silent agreement-by-default: accepting the answer *because* it sounded right, which is exactly Kahneman's \"the way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n\n**Step 6 — Calibrate.** Keep a running log by domain: where does this model, in your hands, reliably get it right, and where has it burned you? High-calibration domains (e.g., boilerplate code, summarizing text you supply, ideation) earn more System-1 trust. Low-calibration domains (exact citations, current events past the training cutoff, precise numbers, niche law/medicine) get mandatory System 2 every time. This mirrors the human rule — trained System 1 is valuable *inside* its calibrated domain and dangerous outside it — now applied to a tool whose calibrated domain you have to learn empirically.\n\n## The Core Lesson\n\nThe AI era does not retire dual-system thinking — it makes it operationally urgent at population scale. The LLM is a spectacularly capable System 1 that you can rent by the token. Its greatest risk is not that it is often wrong, but that it is **confidently, fluently wrong in the same voice it uses when it is right**, stripping away the natural hesitation cues (a human hedging, pausing, saying \"I'm not sure\") that normally prompt us to slow down. The human's irreducible job is to be the System 2: to decide, per the stakes × reliability matrix, when to accept the fast answer and when to force verification — the model's or your own.\n\n*Sources: Kahneman, D. (2011), *Thinking, Fast and Slow* (System 1 / System 2 synthesis; \"feeling right\" vs \"being right\"). Ji, Z. et al. (2023), \"Survey of Hallucination in Natural Language Generation,\" *ACM Computing Surveys* 55(12) — defines and surveys LLM hallucination/confabulation. Widely reported 2023–2024 U.S. legal sanctions over AI-fabricated case citations (e.g., *Mata v. Avianca*, S.D.N.Y., 2023). OpenAI and Anthropic public documentation (2024–2025) on reasoning/extended-thinking modes and tool use as means of eliciting more deliberate, checkable model output.*\n\nFile v1.0.4:skill-card.md\n\n## Description: <br>\nGuides agents to slow down consequential, obvious-feeling decisions by distinguishing fast intuition from deliberate verification and applying structured checks. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and decision-makers use this skill to coach high-stakes or unfamiliar decisions, compare intuitive and deliberate reasoning, and decide when to verify confident AI or human answers. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill can change an agent's reasoning posture by slowing high-stakes decisions and asking structured follow-up questions. <br>\nMitigation: Use it for consequential, unfamiliar, or bias-prone decisions; skip it for routine low-stakes choices or true emergencies that require speed. <br>\nRisk: Decision coaching can still produce misplaced confidence if users treat the framework as a substitute for verification. <br>\nMitigation: Verify load-bearing claims, consult relevant experts or primary sources when stakes are high, and record outcomes for calibration. <br>\n\n\n## Reference(s): <br>\n- [ClawHub Skill Page](https://clawhub.ai/deciqai/skills/dual-system-thinking) <br>\n- [Primary Sources](references/sources.md) <br>\n- [Kahneman 2011 and Tversky-Kahneman Research Example](examples/kahneman-2011-tversky-kahneman-research.md) <br>\n- [LLMs as System 1 AI-Era Example](examples/llms-as-system-1-ai-era-2024-2026.md) <br>\n- [deciqAI Dual-System Thinking Demo](https://www.deciqai.com/c/dual-system-thinking) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Text] <br>\n**Output Format:** [Markdown decision-coaching template and stepwise questions] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [No executable code; may ask the user to pause at WAIT steps before continuing.] <br>\n\n## Skill Version(s): <br>\n1.0.4 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.3: 5 files, 9031 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), references/sources.md (1417b), skill-card.md (2395b), SKILL.md (6934b), _meta.json (139b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast.\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 164 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/dual-system-thinking** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1783508431310\n}\n\nFile v1.0.3:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n\nFile v1.0.3:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.3:skill-card.md\n\n## Description: <br>\nGuides agents through System 1/System 2 checks for consequential decisions, using structured prompts to slow fast intuition, surface possible biases, compare alternatives, and log calibration. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and developers use this skill to structure high-stakes or unfamiliar decisions when fast intuition may need deliberate analysis. It helps an agent ask for the decision context, classify System 1 versus System 2 reasoning, apply a checklist or premortem, compare outputs, and record calibration signals. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Decision worksheets may contain sensitive personal or business details. <br>\nMitigation: Only include details that are appropriate to share with the agent or workspace, and redact unnecessary sensitive information before recording decisions. <br>\nRisk: The skill structures reasoning but does not guarantee correct decisions in consequential contexts. <br>\nMitigation: Use the output as a deliberation aid and review important decisions against evidence, domain expertise, and applicable organizational procedures. <br>\n\n\n## Reference(s): <br>\n- [Sources - dual-system-thinking](artifact/references/sources.md) <br>\n- [Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research](artifact/examples/kahneman-2011-tversky-kahneman-research.md) <br>\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/dual-system-thinking) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Text] <br>\n**Output Format:** [Markdown decision worksheet and coaching prompts] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Includes wait checkpoints and a calibration log template; no code, shell commands, tool calls, persistence, or hidden data access.] <br>\n\n## Skill Version(s): <br>\n1.0.3 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.2: 5 files, 8948 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), references/sources.md (1417b), skill-card.md (2101b), SKILL.md (7044b), _meta.json (139b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast.\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — 163 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/skills/dual-system-thinking?utm_source=clawhub&utm_medium=marketplace&utm_campaign=knowledge-skills&utm_content=dual-system-thinking** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1783471532129\n}\n\nFile v1.0.2:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n\nFile v1.0.2:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.2:skill-card.md\n\n## Description: <br>\nGuides agents through a System 1/System 2 decision check for consequential, unfamiliar, or pressure-driven decisions where fast intuition should be tested with deliberate analysis. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nExternal users and agent builders use this skill to slow down high-stakes or unfamiliar decisions, compare intuitive and analytical outputs, and document a calibrated final decision. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may include sensitive business, medical, legal, or personal decision details in optional decision logs. <br>\nMitigation: Avoid entering sensitive details unless storing or sharing them in the agent environment is acceptable. <br>\nRisk: Cognitive coaching can be mistaken for expert domain judgment in consequential decisions. <br>\nMitigation: Use the skill as decision support and involve qualified reviewers for regulated, professional, or high-impact decisions. <br>\n\n\n## Reference(s): <br>\n- [Sources - dual-system-thinking](references/sources.md) <br>\n- [Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md) <br>\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/dual-system-thinking) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Text] <br>\n**Output Format:** [Markdown decision worksheet and coaching prompts] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May include a decision log template; no executable output.] <br>\n\n## Skill Version(s): <br>\n1.0.2 (source: server release metadata) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.1: 5 files, 8922 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), references/sources.md (1417b), skill-card.md (2242b), SKILL.md (6948b), _meta.json (139b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast.\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with [`metacognition`](../metacognition/SKILL.md), every cognitive-bias skill ([`anchoring`](../anchoring/SKILL.md), [`confirmation-bias`](../confirmation-bias/SKILL.md), [`availability-heuristic`](../availability-heuristic/SKILL.md), etc.), [`deep-work`](../deep-work/SKILL.md), and [`wu-wei`](../wu-wei/SKILL.md).\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.*\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1783456343982\n}\n\nFile v1.0.1:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n\nFile v1.0.1:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.1:skill-card.md\n\n## Description: <br>\nHelps an agent slow down consequential, obvious-feeling decisions by identifying System 1 output, recruiting System 2 analysis, comparing both outputs, and logging calibration signals. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, developers, and agents use this skill to structure high-stakes or unfamiliar decisions that initially feel obvious. It produces a concise decision record that separates fast intuition from deliberate analysis and documents the procedure used. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Decision-support guidance can be misleading if users treat the skill output as authoritative for consequential decisions. <br>\nMitigation: Use the output as a structured thinking aid, review the reasoning, and involve qualified human or domain review for high-stakes decisions. <br>\nRisk: The skill may link to related thinking skills or reference material outside the reviewed artifact set. <br>\nMitigation: Review linked skills and references before relying on them in a deployment. <br>\n\n\n## Reference(s): <br>\n- [ClawHub skill page](https://clawhub.ai/deciqai/skills/dual-system-thinking) <br>\n- [Sources - dual-system-thinking](references/sources.md) <br>\n- [Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Guidance, Markdown, Text] <br>\n**Output Format:** [Markdown decision template and step-by-step coaching prompts] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May include hard-stop prompts for novice coaching and a calibration log template.] <br>\n\n## Skill Version(s): <br>\n1.0.1 (source: server release metadata) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.0: 5 files, 8972 bytes\n\nFiles: examples/kahneman-2011-tversky-kahneman-research.md (6440b), references/sources.md (1417b), skill-card.md (2294b), SKILL.md (6948b), _meta.json (139b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast.\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with [`metacognition`](../metacognition/SKILL.md), every cognitive-bias skill ([`anchoring`](../anchoring/SKILL.md), [`confirmation-bias`](../confirmation-bias/SKILL.md), [`availability-heuristic`](../availability-heuristic/SKILL.md), etc.), [`deep-work`](../deep-work/SKILL.md), and [`wu-wei`](../wu-wei/SKILL.md).\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate + uncertain + felt effort = System 2.\n\n**Step 3 — Recruit decision:**\n\n| Stakes | Familiarity | Recruit System 2? |\n|---|---|---|\n| Low | High | No |\n| Low | Low | Optional |\n| High | High | Yes (verification needed) |\n| High | Low | Mandatory |\n\n**Step 4 — Recruit via procedure:** write down the question; list 3 alternatives; name the active bias; apply a checklist; consult base rates; run a premortem; take 24h if possible.\n\n**Step 5 — Compare outputs:** if S1 and S2 agree → high confidence. If they diverge → S2 wins for novel/high-stakes; in calibrated expert domains, S1 may be right. Document the reason.\n\n**Step 6 — Calibrate:** log decisions + outcomes by domain. High-calibration domains: trust S1 more. Low-calibration: always recruit S2.\n\n## Output Template\n```\nDecision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):\n```\n\n*→ Method in Action: [Kahneman 2011 + 40 Years of Tversky-Kahneman Research](examples/kahneman-2011-tversky-kahneman-research.md)*\n\n## Pack: Key Domain Patterns\n\n| Domain | S1 role | S2 role | Common error |\n|---|---|---|---|\n| Hiring | First impression | Structured rubric; work-sample | Halo effect dominates |\n| Investing | Pattern recognition | Base rates; bear-case | \"Good feeling\" in novel domain |\n| Medical diagnosis | Quick pattern-match | Differential; base rates | Anchoring on first impression |\n| Strategic decisions | Founder intuition | Premortem; market analysis | Confident S1 without calibration |\n\n## Applying It Well\n\n\"Obvious\" is a warning, not a confirmation — treat it as a hypothesis. Structural System 2 recruitment (checklists, premortems, decision journals) beats willpower. Trained System 1 in a calibrated domain is valuable; outside that domain, it's dangerous.\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"I trust my gut\" | Sometimes warranted (calibrated domain). Often a defense against System 2 discipline. Check the track record. |\n| [D] \"It's obvious; we don't need to overthink\" | \"Obvious\" is System 1 output. The check is cheap; the cost of wrong is high. |\n| [D] \"We have to decide fast\" | Often speed pressure is exaggerated. Most \"urgent\" decisions absorb a 1-hour structured pause. |\n| [D] \"I'm an expert; my intuition is reliable\" | Expertise is domain-specific. Rapid, accurate, repeated feedback in this exact domain? |\n| [D] \"Cognitive biases don't apply to me; I know about them\" | Bias-blind-spot effect: knowing reduces biases weakly. Structural procedure is the corrective. |\n| [D] \"The team agrees; we don't need to second-guess\" | Team consensus is often System 1 cascade. Premortems are the System 2 counter. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\nHigh-stakes decision feels obvious and you're not checking it · tired/pressured making consequential decisions · novel domain + fast intuition · team converging without pushback · \"I just know\" is your justification · decision in <30 seconds on meaningful stakes\n\n## Verification\n\n- [ ] System in operation diagnosed (S1 / S2 / mixed)\n- [ ] Stakes × familiarity matrix applied\n- [ ] If recruitment needed, structured procedure used\n- [ ] S1 and S2 outputs compared; divergence justified\n- [ ] Active cognitive biases listed\n- [ ] Decision logged for calibration\n\n---\n\n*Part of **deciqAI Knowledge Skills** — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.*\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1782627440290\n}\n\nFile v1.0.0:references/sources.md\n\n# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n\nFile v1.0.0:examples/kahneman-2011-tversky-kahneman-research.md\n\n# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as automatic, largely unconscious, and relatively undemanding of computational capacity. System 2 processes are characterized as controlled, conscious, and demanding of computational capacity. ... We argue that many of the heuristics and biases studied in the literature reflect default System 1 responses; individual differences in rational thinking reflect differences in the disposition and ability to override these defaults via System 2.\"\n>\n> — Stanovich & West (2000), pp. 645-646.\n\nThe framework has been hugely influential beyond academic psychology:\n\n**Behavioral economics.** The entire field of behavioral economics (Thaler 2017 Nobel; Sunstein's nudge framework) rests on dual-system thinking. Default-setting, choice architecture, framing effects, and other operational interventions all work by leveraging System 1's properties.\n\n**Medical decision-making.** Diagnostic medicine has been profoundly affected. Pattern-recognition (System 1) is the basis of expert diagnosis, but System 1 errors (anchoring on a first diagnosis, availability bias on recently-seen cases) account for ~70% of diagnostic errors. Decision-support systems force System 2 engagement on differential diagnoses.\n\n**Judicial and legal reasoning.** Research on jury decision-making, judicial determinations, and legal-evidence evaluation all uses the dual-system framework. The interventions (structured deliberation, written opinions, appellate review) are explicit System 2 recruitment.\n\n**Marketing and product design.** Dual-system insight underlies modern marketing — System 1 responses to packaging, branding, scarcity signals; System 2 engagement for high-consideration purchases. Conversion-optimization, dark patterns, and consent-flow design all operate on the framework.\n\n**Public policy and choice architecture.** Sunstein and Thaler's \"nudge\" framework operationalizes the dual-system insight: design defaults and choice presentations such that System 1's tendencies produce socially-beneficial outcomes (e.g., opt-out 401k enrollment, organ donation defaults). See [`hyperbolic-discounting`](../skills/hyperbolic-discounting/SKILL.md).\n\n**Executive decision-making.** The Kahneman-cited research on executive over-confidence, planning fallacy, and bias in M&A decisions has shaped how sophisticated decision-makers structure their own processes — premortems, devil's advocates, decision journals, all are explicit System 2 recruitment.\n\nThree operational lessons:\n\n**First, \"obvious\" is a warning, not a confirmation.** When a high-stakes decision feels obvious, that feeling is a System 1 output. Treat it as a hypothesis to be checked, not as a finding. The discipline of \"what would the System 2 analysis show?\" prevents most consequential errors.\n\n**Second, structural System 2 recruitment beats willpower.** You cannot reliably remember to think hard about every decision. The structural interventions (checklists, premortems, decision journals, base-rate consultation, ACH) work because they force System 2 engagement procedurally, not through ongoing vigilance.\n\n**Third, trained System 1 in a calibrated domain is valuable; trained System 1 outside its domain is dangerous.** Expert intuition is real (Klein 1998, *Sources of Power*) — but only in domains with rapid, accurate, repeated feedback. In novel or low-feedback domains, \"trust your gut\" is \"trust System 1 to do something it wasn't trained to do.\" Calibrate.\n\nFile v1.0.0:skill-card.md\n\n## Description: <br>\nGuides an agent through a dual-system decision process that separates fast intuition from deliberate analysis for consequential or unfamiliar choices. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nAgents use this skill to coach users, developers, and teams through high-stakes or unfamiliar decisions by identifying the initial intuitive answer, deciding whether deliberate review is needed, and applying structured checks such as alternatives, checklists, base rates, premortems, and calibration logs. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill can produce decision advice that users may over-trust in medical, legal, financial, or other high-stakes contexts. <br>\nMitigation: Treat outputs as decision-support guidance, require human judgment, and consult qualified professionals for regulated or high-impact decisions. <br>\nRisk: The framework may slow or overcomplicate genuinely routine, low-stakes, or time-critical decisions. <br>\nMitigation: Apply the skill only when the evidence-supported activation conditions are met: consequential, unfamiliar, pressured, depleted, or rapidly converging decisions. <br>\n\n\n## Reference(s): <br>\n- [Primary sources for dual-system thinking](artifact/references/sources.md) <br>\n- [Method in Action: Kahneman 2011 and Tversky-Kahneman Research](artifact/examples/kahneman-2011-tversky-kahneman-research.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown decision-coaching template with structured prompts and a calibration log] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [No executable output; responses may include staged questions that stop for user input.] <br>\n\n## Skill Version(s): <br>\n1.0.0 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>","readmeExcerpt":"Skill: Dual-System Thinking (System 1 / System 2) Owner: deciqai Summary: Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:58:16.166Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/dual-system-thinking.json) v1.0","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"},{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"},{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"},{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"},{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"},{"language":"text","snippet":"Decision: | Initial answer: | Time to arrive: | Confidence:\nSystem in operation: S1 / S2 / mixed\nStakes: H/L | Familiarity: H/L | Recruit S2: Y/N\nProcedure used: | S2 outputs:\nS1 said: | S2 said: | Agree/disagree: | Final answer + reason:\nCalibration log — Domain: | System used: | Outcome (when known):"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: dual-system-thinking\ndescription: \"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast without pushback; you're tired or pressured and still need to decide; you're in an unfamiliar domain moving on instinct. Do NOT activate when: the decision is genuinely routine and low-stakes; you have well-trained calibrated expert intuition in that exact domain and need to act fast. More: deciqai.com/c/dual-system-thinking\"\n---\n\n# Dual-System Thinking (System 1 / System 2)\n\n## Overview\n\nTwo parallel modes: **System 1** — fast, automatic, effortless (pattern recognition, gut feel). **System 2** — slow, deliberate, effortful (analysis, computation, critical evaluation). System 1 runs ~95% of decisions by default; System 2 engages only when recruited. Most cognitive biases are System 1 shortcuts misapplied where System 2 should have intervened.\n\nComposes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.\n\n## When to Use\n\n- Decision feels obvious and is high-stakes; unfamiliar domain + fast intuition; time pressure on consequential matter; team converging without stress-test; depleted (tired/hungry/pressured)\n- Designing a process or interface that depends on user attention\n- Acting on a fast, fluent, confident AI/LLM answer (AI adoption, AI hype, \"the AI said so\") on a high-stakes call — when to force slow verification of a confident machine response\n\n**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** user has a concrete decision → run The Process directly.\n- **Coach mode:** user is unfamiliar → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: when a high-stakes decision feels obvious, that obviousness is a warning — System 1 is firing and may need System 2 to check it.\n2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.\n3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?\n> **[WAIT — do not advance until user responds]**\n4. Work through: what's the System 1 answer? what's the System 2 check? what bias might be active? what procedure forces System 2 engagement?\n> **[WAIT — do not advance until user responds]**\n5. Close: name the system used and the procedure applied.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.\n\n**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate +"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"dual-system-thinking\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1784224696166\n}"},{"path":"references/sources.md","content":"# Sources — dual-system-thinking\n\n> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*\n\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.\n- Tversky, A. & Kahneman, D. (1974). \"Judgment under Uncertainty: Heuristics and Biases.\" *Science*, 185(4157), 1124-1131. The foundational paper.\n- Kahneman, D. & Tversky, A. (1979). \"Prospect Theory: An Analysis of Decision under Risk.\" *Econometrica*, 47(2), 263-291.\n- Stanovich, K. E. & West, R. F. (2000). \"Individual differences in reasoning: Implications for the rationality debate?\" *Behavioral and Brain Sciences*, 23(5), 645-665. The System 1 / System 2 terminology.\n- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.\n- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.\n- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.\n- Evans, J. St. B. T. (2008). \"Dual-processing accounts of reasoning, judgment, and social cognition.\" *Annual Review of Psychology*, 59, 255-278. The academic review.\n- Pronin, E., Lin, D. Y., & Ross, L. (2002). \"The bias blind spot: Perceptions of bias in self versus others.\" *Personality and Social Psychology Bulletin*, 28(3), 369-381.\n- Ji, Z., Lee, N., Frieske, R., et al. (2023). \"Survey of Hallucination in Natural Language Generation.\" *ACM Computing Surveys*, 55(12), 1-38. LLMs as fluent-but-uncalibrated System-1-like responders (2024–2026 AI-era example).\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461, S.D.N.Y. (2023). Attorneys sanctioned for a filing containing AI-generated citations to non-existent cases — a widely reported high-stakes failure of un-verified, confident AI output."},{"path":"examples/kahneman-2011-tversky-kahneman-research.md","content":"# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nThe framework's empirical foundation is the Tversky-Kahneman research program, which began in 1969 at Hebrew University in Jerusalem and continued until Tversky's death in 1996. The program produced ~30 major papers, documenting one cognitive bias after another, each time with the same structure: a specific intuitive judgment that subjects make confidently, but that diverges systematically from rational or statistical analysis.\n\nThe 1974 *Science* paper \"Judgment under Uncertainty: Heuristics and Biases\" articulated three foundational heuristics:\n\n- **Representativeness** — judging probability by similarity to a stereotype (e.g., the \"Linda problem\" — subjects judge Linda more likely to be a \"feminist bank teller\" than a \"bank teller,\" violating basic probability)\n- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))\n- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))\n\nThe 1979 *Econometrica* paper \"Prospect Theory\" articulated the value function and probability-weighting function that account for choice under risk (see [`loss-aversion-prospect-theory`](../skills/loss-aversion-prospect-theory/SKILL.md)).\n\nOver the following 30+ years, dozens of additional biases were documented: framing effects, sunk-cost reasoning, hindsight bias, confirmation bias, the planning fallacy, the conjunction fallacy, the base-rate fallacy, etc.\n\nThe dual-system framework was Kahneman's 2011 synthesis. It answered the question \"why are humans biased in these systematic ways?\" with: *because System 1 produces fast heuristic answers that work in familiar situations and fail in novel/statistical/high-stakes ones, and System 2 is too lazy to override System 1 most of the time.*\n\nKahneman's most operationally cited passage:\n\n> \"A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact. ... Anything that makes it easier for the associative machine [System 1] to run smoothly will also bias beliefs. ... The conclusion is straightforward: the way to be a hero in System 2 is to actively engage when a decision matters. The way to be a victim of System 1 is to assume that 'feeling right' is the same as 'being right.'\"\n>\n> — Kahneman (2011), p. 62.\n\nThe Stanovich-West 2000 paper that named the systems was technical, drawing on individual-difference research showing that people varied in their tendency to engage System 2 and that this tendency correlated with rationality measures:\n\n> \"We use the terms System 1 and System 2 to refer to a broad distinction observed in the literature on dual-process theories. System 1 processes are characterized as aut"},{"path":"examples/llms-as-system-1-ai-era-2024-2026.md","content":"# Method in Action: LLMs as System 1 — When to Force System 2 on a Confident AI Answer (2024–2026)\n\n> *Example for the [dual-system-thinking](../SKILL.md) skill.*\n\nBy 2024–2026, hundreds of millions of people were routing everyday judgments through large language models (ChatGPT, Claude, Gemini, and copilots embedded in search, IDEs, and office software). A useful and increasingly common framing among researchers and practitioners: a modern LLM behaves like an externalized **System 1** — it returns a fast, fluent, confident answer by pattern-completion, with no felt effort and no built-in signal for when it is guessing. The fluency is the trap. A wrong answer and a right answer arrive in the same authoritative prose, at the same speed, with the same surface confidence.\n\nThis example runs that situation through the skill's own Process. The \"decision\" is not \"what did the AI say\" but *\"should I act on this AI answer as-is, or recruit System 2 to verify it?\"*\n\n## The Analogy (and its limits)\n\nAn LLM maps onto **System 1** in the ways that matter operationally: automatic, effortless from the user's side, pattern-driven, fast, and frequently overconfident. It is *not* literally a human System 1, and it can be *made* to do System-2-like work — chain-of-thought prompting, tool use, retrieval, self-critique, and the \"reasoning\" model modes released across 2024–2025 all push the model toward slower, more deliberate, checkable output. The lesson is not \"AI = System 1, full stop.\" It is: **the default, un-recruited mode of an LLM is System-1-shaped, and the human on the other end is the System 2 that has to decide when to recruit deliberation — the model's or their own.**\n\n## The Process\n\n**Step 1 — Identify decision + initial answer.** Decision: *do I ship / send / act on this LLM output?* The \"initial answer\" is the model's response, which arrives in seconds, reads as confident, and produces almost no felt effort in the user. Note the tell: your own confidence tends to be borrowed from the *fluency* of the text, not from any verification you performed.\n\n**Step 2 — Classify system.** The output is quintessential System 1: quick, automatic, confident, effortless. Critically, this holds even when the content is a fabrication. LLMs produce **hallucinations** (also called confabulations) — fluent, plausible, well-formatted statements that are simply false: invented citations, non-existent case law, wrong API signatures, made-up statistics. Fluency is not calibration. Treat the response as a System 1 output regardless of how authoritative it sounds.\n\n**Step 3 — Recruit decision.** Apply the skill's stakes × familiarity matrix — but read \"familiarity\" as *the model's* reliability in this exact domain *and* your own ability to check it:\n\n| Stakes | Domain the AI is reliable in? | Recruit System 2? |\n|---|---|---|\n| Low (draft an email, brainstorm) | High | No — let System 1 run |\n| Low | Low | Optional |\n| High (legal filing, medical, financial, production code, "}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou... Skill: Dual-System Thinking (System 1 / System 2) Owner: deciqai Summary: Activate when: a decision feels obvious but the stakes are high; someone says 'I just trust my gut' on a consequential call; a team is converging fast withou... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:58:16.166Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/dual-system-thinking.json) v1.0","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1989,"uniquenessScore":49,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T06:29:45.704Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-11T08:42:06.100Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}