thinking-steel-manning

Constructs the strongest opposing case before accepting or rejecting a proposal.

Updated May 24, 2026
One-click install
npx skills add https://github.com/MWest2020/skill-forge --skill thinking-steel-manning-mwest2020
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: thinking-steel-manning
Source: https://github.com/MWest2020/skill-forge/tree/main/skills/thinking-steel-manning
Command: npx skills add https://github.com/MWest2020/skill-forge --skill thinking-steel-manning-mwest2020

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It prevents premature rejection of proposals and reflexive agreement by forcing engagement with the strongest faithful version of the opposing argument before a decision is made. ## Core Features & Use Cases - Strongest-countercase construction: Builds the best evidence-based case an informed advocate would make, avoiding caricatures and weak strawmen. - Overturn conditions: Requires a concrete observation that would change the decision, making the reasoning falsifiable. - Structured decision output: Produces a fixed template ending in accept, revise, or reaffirm with explicit residual risks. - Use Case: During an architecture review where you are inclined to reject a proposed caching layer, run the steel-manning procedure to articulate the strongest case for it, state what evidence would overturn your rejection, then reaffirm or update with documented residual risks. ## Quick Start Ask the assistant to steel-man the proposal you are about to reject and state what evidence would change your mind.

Frequently Asked Questions about thinking-steel-manning

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I steel-man an argument before rejecting a proposal?

State the claim in one sentence, extract the legitimate concern behind the opposition, build the strongest evidence-based case an informed advocate would make, then define one observation that would overturn your position. Engage that strong version and decide to accept, revise, or reaffirm.

What is the difference between steel-manning and a pre-mortem?

Steel-manning builds the strongest case for an opposing position to test a decision, while a pre-mortem imagines failure scenarios for a plan you already chose. Use red-team analysis when you need attacker paths instead of opposing arguments.

When should I not use steel-manning?

Skip it for settled facts, security anti-pattern violations, hard requirement breaches, trivial reversible choices, and live emergencies. It is meant for genuine trade-offs, not for defending what is factually wrong or delaying urgent action.

How does steel-manning prevent sycophancy in AI assistants?

When inclined to agree with a user or plan, the procedure forces construction of the best case against that agreement first. The assistant must state an overturn condition and check whether it was looked for before reaffirming.

What output does a steel-manning analysis produce?

It produces a structured record with the claim, core insight, steel-manned argument, overturn condition, response on the merits, a decision of accept, revise, or reaffirm, and explicit residual risks remaining after the decision.