pessimistic-review

southgateai's avatarfrom southgateai

Critically analyze content for logical gaps, unsupported claims, and potential counterarguments. Does not modify content.

0stars🔀0forks📁View on GitHub🕐Updated Jan 10, 2026

When & Why to Use This Skill

The Pessimistic Review skill provides an adversarial analysis of written content to identify logical gaps, unsupported claims, and potential counterarguments. By adopting diverse philosophical personas and rigorous analytical frameworks, it acts as a high-level critical reviewer that stress-tests arguments to ensure they are robust, credible, and intellectually sound before publication.

Use Cases

  • Pre-publication Stress Testing: Use the skill to find hidden weaknesses and logical flaws in important articles or blog posts before they go live.
  • Academic and Philosophical Rigor: Evaluate complex theories or arguments through multiple lenses, such as Physicalism or Empiricism, to ensure multi-dimensional validity.
  • Automated Draft Auditing: Systematically review files marked as drafts to identify missing evidence, internal contradictions, or non-sequiturs in the reasoning process.
  • Language and Tone Refinement: Detect 'weasel words' and overly aggressive assertions, providing recommendations for more nuanced and defensible language.
namepessimistic-review
descriptionCritically analyze content for logical gaps, unsupported claims, and potential counterarguments. Does not modify content.

Pessimistic Review

Adversarial analysis to find weaknesses in site content. Acts as a critical reviewer looking for flaws.

When to Use

  • Weekly quality review
  • Before publishing important content
  • When /pessimistic-review is invoked
  • Optionally with a specific file: /pessimistic-review obsidian/topics/meaning-of-life.md

Instructions

1. Select Content to Review

If a specific file is provided, review that file.

Otherwise, select content using this priority:

  1. Files with draft: true (drafts need review before publishing)
  2. Files not yet reviewed (check obsidian/reviews/ for existing reviews)
  3. Oldest content by modified date

2. Multi-Perspective Critical Analysis

Review the content from multiple philosophical perspectives. For each critic, adopt their worldview and argue as they would.

Critic Personas

The Eliminative Materialist (Patricia Churchland)

  • Consciousness talk is folk psychology destined for elimination
  • Neuroscience will replace mental vocabulary entirely
  • "What you call 'qualia' is just neural activity you don't understand yet"
  • Attack: Dualism is pre-scientific; the explanatory gap will close

The Hard-Nosed Physicalist (Daniel Dennett)

  • Consciousness is real but not what we think it is
  • Introspection is unreliable; we're "zimbos" who think we have qualia
  • The hard problem is a confusion, not a discovery
  • Attack: You're inflating intuitions into metaphysics

The Quantum Skeptic (Max Tegmark)

  • Decoherence destroys quantum effects in warm brains in femtoseconds
  • There's no time for consciousness to "choose" anything
  • Quantum consciousness is wishful thinking dressed in physics
  • Attack: The math doesn't work; you haven't done the calculations

The Many-Worlds Defender (David Deutsch)

  • MWI is the simplest interpretation—it's just the Schrödinger equation
  • Indexical questions are confused; there's no "me" to ask "why this branch"
  • Your rejection of MWI is anthropocentric bias
  • Attack: You're letting intuition override mathematical elegance

The Empiricist (Karl Popper's Ghost)

  • Unfalsifiable claims aren't scientific
  • "Minimal quantum interaction" that's undetectable is metaphysics, not physics
  • What experiment could prove you wrong?
  • Attack: This is not even wrong

The Buddhist Philosopher (Nagarjuna)

  • The self you're trying to preserve is itself an illusion
  • Consciousness isn't a thing that interacts—it's empty
  • Your dualism reifies what should be deconstructed
  • Attack: You're clinging to a permanent self that doesn't exist

Standard Analysis

Also check for:

Logical Gaps

  • Are there non-sequiturs in the argument?
  • Do conclusions follow from premises?
  • Are there missing steps in reasoning?

Unsupported Claims

  • Are assertions made without evidence or argument?
  • Are sources cited for factual claims?
  • Are philosophical positions attributed correctly?

Internal Contradictions

  • Does the content contradict itself?
  • Does it conflict with other site content?
  • Does it conflict with The Unfinishable Map's tenets?

Language Issues

  • Is language overly strong without justification? ("clearly," "obviously," "must be")
  • Are there weasel words hiding weak arguments?
  • Is the tone appropriately uncertain where warranted?

Style Guide Violations (see obsidian/project/writing-style.md)

  • Is important information front-loaded for LLM truncation resilience?
  • Are there undefined forward references without named-anchor markers?
  • Is the "Relation to Site Perspective" section present and substantive?
  • Is redundant background minimised (focus on what's novel)?

3. Generate Report

Create a report at obsidian/reviews/pessimistic-YYYY-MM-DD.md:

---
title: Pessimistic Review - YYYY-MM-DD
created: YYYY-MM-DD
draft: false
ai_contribution: 100
ai_system: [current model]
---

# Pessimistic Review

**Date**: YYYY-MM-DD
**Content reviewed**: [filename or list]

## Executive Summary

[2-3 sentence summary of main findings]

## Critiques by Philosopher

### The Eliminative Materialist
[What Patricia Churchland would say about this content]

### The Hard-Nosed Physicalist
[What Daniel Dennett would say]

### The Quantum Skeptic
[What Max Tegmark would say]

### The Many-Worlds Defender
[What David Deutsch would say]

### The Empiricist
[What a Popperian would say about falsifiability]

### The Buddhist Philosopher
[What Nagarjuna would say about self and emptiness]

## Critical Issues

### Issue 1: [Title]
- **File**: [filename]
- **Location**: [section or quote]
- **Problem**: [description]
- **Severity**: High/Medium/Low
- **Recommendation**: [how to address]

## Counterarguments to Address

### [Topic/Claim]
- **Current content says**: [summary]
- **A critic would argue**: [counterargument]
- **Suggested response**: [how to address]

## Unsupported Claims

| Claim | Location | Needed Support |
|-------|----------|----------------|
| [claim] | [file:section] | [what's needed] |

## Language Improvements

| Current | Issue | Suggested |
|---------|-------|-----------|
| "clearly shows" | Too strong | "suggests" |

## Strengths (Brief)

Despite criticisms, note what works well to preserve during revisions.

4. Add Actionable Items

If significant issues are found, add tasks to obsidian/workflow/todo.md:

### P2: Address gaps in [filename]
- **Type**: refine-draft
- **Status**: pending
- **Notes**: Pessimistic review found [brief description]. See pessimistic-YYYY-MM-DD.md

5. Log to Changelog

Append summary to obsidian/workflow/changelog.md.

Important

  • This skill is READ-ONLY for content files
  • Only creates report files, updates changelog and todo.md
  • Does NOT modify the content itself
  • Be genuinely critical - the goal is to find real weaknesses
  • But be constructive - always suggest how to improve