Stats
Actions
Tags
From skills-for-humanity
Assess the fast-thinking pattern at work — when it's reliable, when it misleads, and whether to trust or override it.
How this skill is triggered — by the user, by Claude, or both
Slash command
/skills-for-humanity:s4h-psychology-heuristicsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Fast thinking is not sloppy thinking — it's compressed expertise. Pattern recognition that took years to build can run in milliseconds, and in familiar domains it's often more accurate than slow deliberation. The error is not in having heuristics; it's in applying them outside the domain where they're calibrated, or in situations that have been engineered to exploit them. The question is never ...
Fast thinking is not sloppy thinking — it's compressed expertise. Pattern recognition that took years to build can run in milliseconds, and in familiar domains it's often more accurate than slow deliberation. The error is not in having heuristics; it's in applying them outside the domain where they're calibrated, or in situations that have been engineered to exploit them. The question is never "did I use a heuristic?" (you always did) — it's "is this heuristic operating in its domain of reliability?"
Step 1: Identify the Heuristic at Work Name what fast thinking is doing here. Common heuristics:
Framing check: Confirm the specific judgment or decision before continuing. State what you've identified — the actual situation where fast thinking is being used and what is at stake — in one sentence, then use AskUserQuestion:
Question: "I'm reading this as: [your one-sentence framing of the specific situation and the heuristic potentially at work]. Is that right?"
Header: "Framing"
Options:
Representativeness — Judging probability by how much something resembles the prototype of a category. "This startup's pitch sounds like every successful startup I've seen." Fast and often right within familiar patterns; fails when base rates matter (most startups fail regardless of how compelling the pitch sounds).
Availability — Judging frequency or likelihood by how easily examples come to mind. Recent, vivid, or emotionally charged examples feel more probable. Fails when the most available examples are systematically unrepresentative (media coverage, personal experience).
Affect heuristic — If you feel good about something, you perceive it as lower risk and higher benefit; if you feel bad, higher risk and lower benefit. Fast integration of complex information; fails when the feeling is a response to something unrelated to the actual decision.
Recognition heuristic — Preferring the recognized option when one option is recognized and another isn't. "I've heard of this company, so it must be better." Adaptive when recognition correlates with quality; fails when recognition is driven by marketing rather than merit.
Fluency heuristic — Judging things that are easier to process as more true, more valuable, or more trustworthy. Clear writing, simple numbers, and familiar ideas benefit from this; it penalizes novelty and complexity that is genuine.
Social consensus — Using what others are doing as a guide to what's correct. Adaptive in stable environments with accumulated collective wisdom; fails in novel situations, bubbles, or when the crowd is itself reacting to a cascade.
Expert intuition — Pattern recognition built through deliberate practice in a domain with regular feedback. Reliable in high-validity environments (chess, firefighting, intensive care); unreliable in low-validity environments where feedback is delayed, noisy, or absent (financial forecasting, hiring decisions).
Step 2: Classify the Domain Is the heuristic being applied in its domain of reliability? The key dimensions:
Step 3: Assess Systematic Distortion Risk Is there a predictable direction in which this heuristic, in this context, is likely to mislead?
Step 4: Decide Whether to Trust, Override, or Supplement
Trust the fast thinking when:
Override and apply System 2 analysis when:
Supplement (use both) when:
Before proceeding, use the AskUserQuestion tool. State your interpretation of the situation in 1–2 sentences — what is being analyzed and what the core question is — then ask:
Proceed based on their selection. If the user reframes, incorporate the correction before running any analysis.
[Restate the decision or judgment being made in one sentence]
[Heuristic name] — [How it's operating in this specific situation]
Direction: [Which way does this heuristic likely err in this context?] Magnitude: [Rough sense of how far off it might be]
[Trust / Override / Supplement] — [One paragraph explaining why, and what to do]
The goal is calibration, not skepticism. Dismissing all heuristics produces analysis paralysis; the research on expert intuition shows that in high-validity domains, fast thinking from genuine experts outperforms deliberate analysis. Klein's research on naturalistic decision-making, and the disagreement between him and Kahneman, is instructive: they are both right about different domains.
Use psychology-cognitive-biases when the question is about systematic distortions in a group's beliefs or decisions (biases operate at the population level and compound over time). Use psychology-heuristics when the question is about a specific instance of fast thinking and whether to rely on it. They overlap but have different entry points: biases are about distorted conclusions; heuristics are about the reasoning shortcuts that get you there.
After delivering this output, use AskUserQuestion to offer the next move:
/s4h-decision-criteria-weighting — Weight criteria using the right heuristics/s4h-logic-check — Validate where heuristics may mislead the reasoning/s4h-probability-confidence-calibration — Calibrate confidence adjusting for heuristic errorsnpx claudepluginhub human-avatar/skills-for-humanityDiagnoses which cognitive biases are actively distorting thinking in a specific situation. Activates on queries about bias, objectivity, and decision clarity.
Detects and removes cognitive biases from reasoning using Julia Galef's Scout Mindset framework. Provides reversal tests, scope sensitivity checks, status quo bias tests, confidence interval audits, and full bias audits.
Surfaces 3-4 cognitive biases in prior conversation reasoning or Libertee methods (Six Hats, Debate). Maps biases to session moments and poses one uncomfortable question challenging conclusions. Use after synthesis.