fainshare

fainshare/Liam-Skills

Skill Claude CodeCodex

Verify（验证·太虚三转·叩实）: decompose conclusion → testable predictions → external data validation. Use when user says "验证", "贝叶斯", "bayes", "靠谱吗", "深挖", "/bayes".

fainshare/Liam-Skills

Skill Claude CodeCodex

A method for rebuilding the strongest case for an idea that has received too much criticism.

fainshare/Liam-Skills

Skill Claude CodeCodex

Audit（审查·太虚四转·破妄）: six self-deception principles via blind sub-agent. Use when user says "审查", "费曼", "feynman", "提交前把关", "反自欺", "/feynman".

fainshare/Liam-Skills

Skill Claude CodeCodex

An iterative analysis method that repeats a linked chain of reasoning until the result becomes stable.

fainshare/Liam-Skills

Skill Claude CodeCodex

A divergent-thinking method based on Osborn’s idea-generation techniques, including SCAMPER and rotating through stakeholder viewpoints.

fainshare/Liam-Skills

Skill Claude CodeCodex

A review method for checking the quality of an outside argument through fact-checking, comparison of both sides, and checking whether the evidence is independent. It is different from Feynman, which examines your own reasoning.

fainshare/Liam-Skills

Skill Claude CodeCodex

A structured debate process using five roles across four rounds: idea generation, opposing-view comparison, probability and evidence checks, clear explanation, and review of alternatives. It produces key points, areas of agreement, disagreements, unresolved questions, and an overall judgment.

fainshare/Liam-Skills

Skill Claude CodeCodex

A workflow that runs five thinking methods in sequence: Osborn for generating ideas, Hegel for comparing opposing views, Bayes for updating beliefs with evidence, Feynman for explaining reasoning clearly, and de Bono for examining alternatives.