present-and-defend

Re-present multi-question decision walkthroughs with per-option steel-man defenses, honest weaknesses, and synthesized recommendations.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill present-and-defend-gozonerd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: present-and-defend
Source: https://github.com/gozonerd/shadow-ai-assessment/tree/main/.claude/skills/present-and-defend
Command: npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill present-and-defend-gozonerd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When a multi-question decision walkthrough has been surfaced but the decision is genuinely unclear, a concise "framing + options + recommendation" format lacks the depth needed to weigh tradeoffs. This Skill re-issues the walkthrough with full per-option rationale so the decision-maker can evaluate every choice on its strongest merits before locking. ## Core Features & Use Cases - Steel-Man Defenses: Every option receives its strongest honest defense with evidence and precedent references, never a strawman. - Honest Weakness Sections: Each option's real gaps and failure modes are admitted without false balance or contrived symmetry. - Synthesis with Strength-of-Lean: Per-question synthesis ties recommendations to established discipline patterns and marks each as strongly held, lightly held, or genuinely uncertain. - Use Case: After a walkthrough on architecture migration options is presented and the user says "I genuinely don't know what to choose," invoke this Skill to re-present all options with full defenses, weaknesses, and a summary recommendation table. ## Quick Start Invoke the skill by saying "present and defend the options from the last walkthrough" after a multi-question decision has been surfaced but not yet locked.

Frequently Asked Questions about present-and-defend

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

FAQPage Schema
How do I get deeper rationale for a multi-option decision?

Invoke the present-and-defend skill after a decision walkthrough has been surfaced. It re-presents every option with a full steel-man defense, an honest weakness section, and a per-question synthesis with a strength-of-lean rating.

What is steel-manning in decision analysis?

Steel-manning means constructing the strongest possible honest argument for each option rather than a weak version easy to dismiss. This skill requires every option's defense to cite specific evidence, precedents, or established patterns before admitting its weaknesses.

When should I not use present-and-defend?

Skip it when the decision is already locked, when the question is a trivial yes/no choice, or when the user explicitly delegates the decision. Re-presenting after a lock wastes time, and deep rationale adds no value to single-axis questions.

Does present-and-defend drop options that seem obviously wrong?

No. The skill re-presents all options from the original walkthrough without narrowing. The decision-maker chooses; the skill's job is to defend every option honestly, including ones that appear weaker.

How does the skill avoid false balance between options?

Weakness sections must name real gaps, not contrived ones manufactured to make options look comparable. If one option has a fatal flaw it is named as fatal, and the synthesis reflects genuine asymmetry between options.