What makes Dual-System Thinking (System 1 / System 2) notable?

1.1K downloads. Trust evidence available

How should Dual-System Thinking (System 1 / System 2) be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Dual-System Thinking (System 1 / System 2)

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

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.1k

Updated

Oct 11, 2026

Version

1.0.5

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.1K downloadsadoption · observed Oct 11, 2026
Latest release: 1.0.5release · observed Jul 16, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:dual-system-thinking

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: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-deciqai-dual-system-thinking/snapshot"

Documentation

CLAWHUB

122,131 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: dual-system-thinking
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 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"
---

# Dual-System Thinking (System 1 / System 2)

## Overview

Two 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.

Composes with `metacognition`, every cognitive-bias skill (`anchoring`, `confirmation-bias`, `availability-heuristic`, etc.), `deep-work`, and `wu-wei`.

## When to Use

- 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)
- Designing a process or interface that depends on user attention
- 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

**Not when:** routine low-stakes decisions; calibrated expert System 1 in this exact domain; emergency requiring speed.

## Coaching Novices (Adaptive Front Door)

- **Engine mode:** user has a concrete decision → run The Process directly.
- **Coach mode:** user is unfamiliar → guide step by step.

In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.

1. 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.
2. Check fit: routine low-stakes decisions are fine for System 1. This framework is for consequential, novel, or statistically-loaded decisions.
3. Elicit the specific decision: what's being decided? Does it feel obvious? Are you tired / pressured / in your area of trained expertise?
> **[WAIT — do not advance until user responds]**
4. 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?
> **[WAIT — do not advance until user responds]**
5. Close: name the system used and the procedure applied.
> **[WAIT — do not advance until user responds]**

## The Process

**Step 1 — Identify decision + initial answer:** note what comes to mind quickly, your confidence, and time-to-answer.

**Step 2 — Classify system:** quick + automatic + confident + no felt effort = System 1; slow + deliberate +

_meta.json

{
  "ownerId": "kn754b8sk22s8c6gjxt02bftbn88q7ye",
  "slug": "dual-system-thinking",
  "version": "1.0.5",
  "publishedAt": 1784224696166
}

references/sources.md

# Sources — dual-system-thinking

> *Primary sources for the [dual-system-thinking](../SKILL.md) skill.*

- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. The synthesis.
- Tversky, A. & Kahneman, D. (1974). "Judgment under Uncertainty: Heuristics and Biases." *Science*, 185(4157), 1124-1131. The foundational paper.
- Kahneman, D. & Tversky, A. (1979). "Prospect Theory: An Analysis of Decision under Risk." *Econometrica*, 47(2), 263-291.
- 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.
- Stanovich, K. E. (2011). *Rationality and the Reflective Mind.* Oxford University Press. ISBN 978-0195341140.
- Klein, G. (1998). *Sources of Power: How People Make Decisions.* MIT Press. ISBN 978-0262611466. Expert intuition / trained System 1.
- Thaler, R. H. & Sunstein, C. R. (2008). *Nudge.* Yale University Press. ISBN 978-0300122237. Choice-architecture applications.
- 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.
- 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.
- 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).
- *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.

examples/kahneman-2011-tversky-kahneman-research.md

# Method in Action: Kahneman 2011 + 40 Years of Tversky-Kahneman Research

> *Example for the [dual-system-thinking](../SKILL.md) skill.*

The 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.

The 1974 *Science* paper "Judgment under Uncertainty: Heuristics and Biases" articulated three foundational heuristics:

- **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)
- **Availability** — judging frequency by ease of recall (see [`availability-heuristic`](../skills/availability-heuristic/SKILL.md))
- **Anchoring and adjustment** — being unduly influenced by an initial number (see [`anchoring`](../skills/anchoring/SKILL.md))

The 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)).

Over 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.

The 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.*

Kahneman's most operationally cited passage:

> "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.'"
>
> — Kahneman (2011), p. 62.

The 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:

> "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

examples/llms-as-system-1-ai-era-2024-2026.md

# Method in Action: LLMs as System 1 — When to Force System 2 on a Confident AI Answer (2024–2026)

> *Example for the [dual-system-thinking](../SKILL.md) skill.*

By 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.

This 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?"*

## The Analogy (and its limits)

An 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.**

## The Process

**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.

**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.

**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:

| Stakes | Domain the AI is reliable in? | Recruit System 2? |
|---|---|---|
| Low (draft an email, brainstorm) | High | No — let System 1 run |
| Low | Low | Optional |
| High (legal filing, medical, financial, production code,

Github ReposUpdated 2d agoRank 70

AionUi

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!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

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

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.