Skip to content
Automation
Skill

/construct-critique

Construct the strongest structured attack on a claim/artifact, grounded in explicit reasons/evidence.

From plugin
de-anthropocentric-research-engine
499200 skills
Install
$ npx -y skills add yogsoth-ai/de-anthropocentric-research-engine --skill construct-critique --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/construct-critique

Context preview

The summary Claude sees to decide when to auto-load this skill.

Construct the strongest structured attack on a claim/artifact, grounded in explicit reasons/evidence.

SKILL.md

construct-critique.SKILL.md
name: construct-critique
description: "Construct the strongest structured attack on a claim/artifact, grounded in explicit reasons/evidence."

construct-critique

Purpose

Construct the strongest structured attack on a claim or artifact from an explicit perspective, grounded in reasons and evidence.

Input contract

required: [target_claim_or_artifact, critique_perspective, evidence_set]
optional: [argument_schema, known_assumptions, decision_context]
constraints: [separate claim, ground, warrant, rebuttal, and severity; critique must target the supplied artifact]

Procedure

1. State the target claim and its intended scope. 2. Generate attacks from the supplied perspective using the argument schema. 3. Attach evidence and classify each attack as formal, empirical, scope, or implementation-related. 4. Rank severity, identify the strongest-point attack, and return rebuttal requirements.

Output contract

produces: [critique_set, evidence_links, severity_labels, strongest_attack, rebuttal_requirements]
delta_fields: [findings, evidence_updates, uncertainties, recommended_jumps]

Quality gates

  • In critic-attack mode, produce at least 3 attacks, label each HIGH/MEDIUM/LOW, attach evidence, and include at least 1 attack against the strongest point.
  • Toulmin critiques preserve claim, ground, warrant, and rebuttal fields.
  • Perspective changes the attack lens, not the target claim or evidence standard.

Parameterization

The caller must provide target schema, critique perspective(s), evidence set, argument model, severity scale, and whether the task is debate, convergence, or perspective review.

Failure and counterexamples

Reject attacks that merely restate disagreement without a reason or evidence, and reject severity labels without a stated scale.

Provenance map

  • resolved: stress-test/debate-critic
  • resolved: convergence/critic-attack
  • resolved: stress-test/perspective-critic

Preserved source criteria ledger

| source | physical line | kind | source criterion | |---|---:|---|---| | convergence/critic-attack | 27 | numeric/gate | Must produce >=3 attacks, each with HIGH/MEDIUM/LOW severity and evidence, including at least 1 attack against the strongest point. |

Read more
Ships withde-anthropocentric-research-engine

The complete research orchestration system for AI-native science. What It Does Design Philosophy Architecture (v3.2.2) Quick Start Configuration Roadmap License DARE is not a tool that helps you do research. It is the researcher.

Get the whole plugin