Free portable Agent Skill

Metacognition

Review how a judgment was formed without confusing confidence, outcome and hindsight with decision quality. Use for retrospectives, decision journals and learning from past choices.

Agent SkillsHermes AgentOpenClawChatGPT SkillsCodex

Open SKILL.md Back to marketplace

What it does

A repeatable workflow, not another bias-name list.

Use it when

A good result is being treated as proof that the original decision process was good.
A bad result makes earlier warning signs feel obvious in retrospect.
A review is mainly testing arguments that support the preferred story.
A person or team wants to learn from a decision without rewriting what was knowable at the time.

It returns

What was knowable then
Decision process vs outcome
Original assumptions / confidence
Alternative explanations
Learning that transfers
Change for the next decision

Install

One SKILL.md. No API key.

Hermes Agent

mkdir -p ~/.hermes/skills/metacognition && curl -fsSL https://cognitive-biases.github.io/agent-skills/metacognition/SKILL.md -o ~/.hermes/skills/metacognition/SKILL.md

OpenClaw

mkdir -p ~/.openclaw/workspace/skills/metacognition && curl -fsSL https://cognitive-biases.github.io/agent-skills/metacognition/SKILL.md -o ~/.openclaw/workspace/skills/metacognition/SKILL.md

Other Agent Skills-compatible tools

Download SKILL.md and place it in the skill directory supported by your agent. ChatGPT Skills and Codex support the Agent Skills format; their installation surface can vary by product.

Safety boundary

Useful skepticism without amateur diagnosis.

Do not use the outcome alone to grade the original decision.
Do not claim warning signs were obvious without contemporaneous evidence.
Do not turn the review into a personality judgment.
Distinguish a repeatable process lesson from a one-off story.

Instruction-only · no secrets · no executable code · no required network access · CC BY-NC-SA 4.0

Evidence-linked lenses

The skill checks patterns. It does not diagnose people.

Outcome Bias

Outcome bias occurs when knowledge of a result changes how people evaluate the quality of a decision even when the information available at the time of the decision is held constant. Outcomes can still be relevant for learning, so the error is not 'never look at results'; it is using luck or hindsight as if it had been available to the original decision-maker.

Hindsight Bias

Knowing an outcome can make the outcome look more predictable in retrospect. The effect has been studied for decades and across many settings, but it does not mean that every confident explanation after an event is biased.

Confirmation Bias

Confirmation bias is an umbrella label for several ways existing beliefs or hypotheses can influence information search and interpretation. It should not be reduced to one behaviour such as reading only agreeable news, and a preference for confirming tests is not irrational in every task or environment.

Moral Luck

Resultant moral luck describes cases where judgments of blame, punishment, or moral evaluation differ because otherwise similar actions lead to different outcomes partly outside the agent's control. Outcome information does affect moral judgment in experiments, but the effect should not be reduced to 'people ignore intent.' Mental states, causal responsibility, belief justification, negligence, and the kind of moral judgment being asked about all matter. Some studies find that false or unjustified beliefs explain more of classic moral-luck asymmetries than the bad outcome itself, while still detecting an independent outcome effect.