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
507200 skills

Claude Code

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

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
@yogsoth-ai
Ships withde-anthropocentric-research-engine

A research skill graph for AI-native science. 271 markdown files. No runtime, no dependencies, no build step.

Stars on GitHub

Momentum

stars in total
507
Star history of yogsoth-ai/de-anthropocentric-research-engine on GitHub, a chart by star-history.com

More skills from other plugins.

agent-builder

agent-builder

Load immediately after an Agent intent, before the first agent_builder_* call. Governs how you create, configure, test, and…

n8n11d209k
@n8n-io@n8n-ioInstall Skill
browser-use

browser-use

Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers…

@browser-use@browser-useInstall Skill
agent-browser

agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling…

@vercel-labs-2@vercel-labs-2Install Skill
add-anydoc

add-anydoc

Add local office-document-to-Markdown conversion to NanoClaw agent containers with the pinned Firecrawl AnyDoc CLI. Use when…

@nanocoai@nanocoaiInstall Skill

Questions about construct-critique.

Which coding agents can use construct-critique?
construct-critique installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.