AI Agent Hub
Back to skills
Metacognition Coach icon

Metacognition Coach

Life Service Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_31c5e2a9/meta-coach.

About this skill

The Problem

Once a person has formed a conclusion, the usual risk is not lack of thought but biased reasoning: collecting only evidence that supports the view, treating “everyone thinks so” as fact, weighing losses more than gains, or mixing short-term benefits with long-term costs. This skill targets those pre-decision friction points by using follow-up questions to expose reasoning gaps rather than handing over a final answer.

How It Works

After coach or metacognition is entered, the skill works through several structured checks:

  • Counter-evidence review: ask for at least three pieces of evidence against the current conclusion, identify the strongest objection, and test whether the conclusion could be reversed.
  • Consequence calibration: probe whether the worst outcome is truly acceptable, which option would cause regret in one year, and whether choosing A and failing would trigger regret about not choosing B.
  • Bias checks: surface one-sided evidence collection, make “everyone” concrete, and compare what is lost versus what is gained.
  • Retrospective simulation: ask what the largest gap between outcome and expectation would be, where the cause appears, and how the next decision point should change.
  • Contradiction detection: catch inconsistent motives, such as wanting to stay still while giving two conflicting reasons.

It fits product decisions, technical trade-offs, career choices, relationship judgments, and other scenarios where self-reasoning needs calibration.

Boundaries

This is a conversational probing skill. It does not make the final decision for the user and should not be treated as legal, medical, or financial advice. For fact checking, external data validation, or tasks requiring a formal evidence chain, it can only help clarify thinking, not replace research, analysis, or professional consultation.

Use Cases

  • Before product review, ask coach to list three objections and identify strongest evidence against shipping.
  • When choosing between technical options, probe worst-case acceptability, one-year regret, and cost of failed choice A.
  • Before final report, challenge vague 'everyone' claims and unbalanced loss/gain framing in the argument.
  • During postmortem, surface contradictory reasons, expectation gap, and next decision-point changes.

Best For

  • Product managers who need pre-launch conclusion checks and want counter-evidence, not only support.
  • Backend engineers making architecture trade-offs and needing bias and regret-cost review.
  • Tech leads writing proposals who need concrete 'everyone' claims and balanced loss/gain framing.
  • Career or project decision-makers who want to inspect expectation gaps and contradictory motives.