Key Features - Argument Cartographer

Interactive Argument Mapping

At the heart of Argument Cartographer is a powerful visualization engine that transforms abstract debates into tangible, navigable maps.

Flow Map

Classical argument structure with thesis at top, reasons and objections flowing down to conclusion

Tree View

Hierarchical tree structure showing parent-child relationships between nodes

Compass View

Circular layout with thesis at center, claims radiating outward by side

Circular View

3D flip cards arranged in circular pattern for engaging exploration

Pillar View

Side-by-side comparison of “For” and “Against” arguments

Balanced View

Split layout showing supporting vs. opposing evidence

Each argument component is represented as an interactive node:

interface ArgumentNode {
  id: string;
  type: 'thesis' | 'claim' | 'counterclaim' | 'evidence';
  side: 'for' | 'against';
  content: string;
  sourceText: string; // Original quote from source
  source: string; // URL
  fallacies: string[];
  logicalRole: string;
}

Clicking any node reveals its full context, source attribution, and detected fallacies in a detailed side panel.

Logical Fallacy Detection

Our AI actively hunts for rhetorical manipulation and logical errors in source material.

Detected Fallacy Types

Ad Hominem - Attacking the Person

Definition: Attacking the character or circumstances of an individual instead of addressing their argument.Example: “You can’t trust her climate change research - she drives an SUV!”Why Problematic: The validity of an argument stands independent of who makes it.

Straw Man - Misrepresenting Opposition

Definition: Distorting or oversimplifying an opponent’s position to make it easier to attack.Example: “Gun control advocates want to ban all guns and leave citizens defenseless.”Why Problematic: Refuting a distorted version doesn’t address the actual argument.

False Dichotomy - Limited Options

Definition: Presenting only two options when more alternatives exist.Example: “Either we cut all social programs or the economy will collapse.”Why Problematic: Oversimplifies complex issues and excludes middle-ground solutions.

Slippery Slope - Chain Reaction

Definition: Assuming a chain of events without evidence that one will lead to another.Example: “If we allow same-sex marriage, next people will want to marry animals.”Why Problematic: Assumes causation without supporting evidence.

Appeal to Authority - False Expertise

Definition: Using an authority figure’s opinion as evidence when they lack relevant expertise.Example: “Einstein believed in God, so atheism must be wrong.”Why Problematic: Expertise in one domain doesn’t transfer to all domains.

Hasty Generalization - Insufficient Evidence

Definition: Drawing broad conclusions from limited or unrepresentative samples.Example: “I met two rude people from that country - everyone there must be rude.”Why Problematic: Small samples can’t support universal claims.

Fallacy Card Interface

Each detected fallacy is presented in an expandable card showing:

Severity Badge

Critical (red), Major (orange), or Minor (yellow) severity indicator

Category Tag

Logical, Rhetorical, or Statistical fallacy classification

Problematic Text

Exact quote from source containing the fallacy

Explanation

Clear explanation of why this is problematic reasoning

Educational Content

Formal definition and how to avoid this fallacy

Suggested Improvement

Logically sound rephrasing of the argument

interface DetectedFallacy {
  id: string;
  name: string;
  severity: 'Critical' | 'Major' | 'Minor';
  category: string;
  confidence: number; // 0-1 AI confidence score
  problematicText: string;
  explanation: string;
  definition: string;
  avoidance: string;
  example: string;
  suggestion: string;
  location?: string; // Where in argument structure
}

Narrative Radar

A real-time “Head-Up Display” for trending topics and breaking news analysis.

How Radar Works

Topic Curation

Editors select high-impact, trending topics from news cycles

Pre-Analysis

AI generates comprehensive blueprints before misinformation spreads

Live Updates

Analyses refresh as new sources and social data emerge

Public Access

All users can explore pre-generated analyses without API costs

Narrative Radar topics are selected based on public interest, social velocity, and potential for misinformation spread.

Radar Features

Trending Topics Feed - Curated list of analyzed debates
Impact Metrics - Social engagement and source count
Freshness Indicators - Last updated timestamps
Quick Preview - Summary and credibility score at a glance

Credibility Scoring

Every analysis receives a “brutally honest” credibility score (1-10) based on multiple factors.

Scoring Algorithm

Source Quality (30%)
Evidence Strength (30%)
Fallacy Presence (30%)
Logical Coherence (10%)

Factors:

Number of independent sources (more is better)
Domain diversity (avoid echo chambers)
Trusted outlet presence (Reuters, BBC, etc.)
Recency of sources

Scoring:

8+ diverse sources from trusted outlets: +3 points
4-7 sources with some diversity: +2 points
1-3 sources or low diversity: +1 point

Score Interpretation

Score	Interpretation	Typical Characteristics
9-10	Exceptional	Multiple primary sources, zero fallacies, strong evidence
7-8	Strong	Good source diversity, minimal fallacies, solid evidence
5-6	Moderate	Some sources, few fallacies, mixed evidence quality
3-4	Weak	Limited sources, several fallacies, opinion-heavy
1-2	Very Weak	Poor sourcing, critical fallacies, unreliable

Even well-argued topics rarely score above 8/10. This is intentional - most real-world debates have legitimate complexity and imperfect evidence.

Real-time integration with Twitter/X to capture public sentiment and discourse.

Data Collection

For each topic, we fetch:

20 most relevant tweets (sorted by engagement)
Author information (name, handle, profile image)
Engagement metrics (likes, retweets, replies, impressions)
Timestamp (when posted)

const searchParams = new URLSearchParams({
  'query': `${topic} lang:en -is:retweet`,
  'tweet.fields': 'created_at,author_id,public_metrics',
  'expansions': 'author_id',
  'user.fields': 'profile_image_url,username,name',
  'max_results': '20',
  'sort_order': 'relevancy'
});

AI-Generated Summary

Our AI analyzes collected tweets and generates a neutral summary capturing:

Dominant sentiment (positive/negative/mixed)
Key discussion points (what people are actually talking about)
Emerging themes (patterns in public discourse)
Notable perspectives (unique or influential viewpoints)

The summary style mimics Twitter/X’s “Grok” feature - concise, neutral, and focused on themes rather than individual opinions.

Tweet Display

Tweets are displayed in interactive cards showing:

Full tweet text with proper formatting
Author name, handle, and avatar
Engagement metrics (visual indicators)
Link to original tweet
Timestamp (relative: “2 hours ago”)

Comprehensive export capabilities for all analysis outputs.

Export Formats

PNG Image

High-resolution raster image

Resolutions: 1x, 2x, 3x
Light or dark theme
Optional background

SVG Vector

Scalable vector graphics

Infinite resolution
Editable in design tools
Small file size

JSON Data

Raw structured data

Full blueprint schema
Fallacy details
Source metadata

Export Options

Visual Settings
Actions

Color Mode: Light or Dark theme
Background: Transparent or themed
Resolution: 1x (standard), 2x (HD), 3x (print-quality)

// PNG export with custom resolution
const dataUrl = await htmlToImage.toPng(element, {
  pixelRatio: 3, // 3x resolution
  backgroundColor: isDark ? 'hsl(217, 14%, 10%)' : 'hsl(40, 50%, 98%)',
});

// SVG export (infinite resolution)
const svgDataUrl = await htmlToImage.toSvg(element, options);

// JSON export (raw data)
const jsonString = JSON.stringify(blueprint, null, 2);

AI-Powered “Ask More”

Interactive chat interface to ask follow-up questions about the analysis.

How It Works

User asks a question about the topic
System provides full blueprint as context to AI
Gemini generates informed response based on analyzed data
Response includes citations to specific blueprint nodes

Ask More is perfect for exploring specific claims, requesting clarification on fallacies, or diving deeper into particular evidence.

Example Questions

“What’s the strongest evidence for the opposing side?”
“Can you explain the Ad Hominem fallacy in claim #3?”
“Which sources are most credible on this topic?”
“What are the weakest arguments on each side?”

Real-Time Collaboration

While currently single-user focused, the architecture supports future collaborative features:

Shared analyses - Public URLs for specific blueprints
Annotation layers - Add personal notes to nodes
Version history - Track how analyses evolve over time
Workspace teams - Collaborative analysis for research teams

Collaboration features are on the roadmap for future releases.

Next Steps

Quick Start

Create your first analysis in 5 minutes

Feature Deep Dives

Explore each feature in detail

Visualization Modes

Master all 6 visualization options

Fallacy Detection

Learn about logical fallacy identification

Overview

Getting Started

Core Features

User Guide

Architecture

​Interactive Argument Mapping

Flow Map

Tree View

Compass View

Circular View

Pillar View

Balanced View

​Node-Based Navigation

​Logical Fallacy Detection

​Detected Fallacy Types

​Fallacy Card Interface

​Narrative Radar

​How Radar Works

​Radar Features

​Credibility Scoring

​Scoring Algorithm

​Score Interpretation

​Social Pulse

​Data Collection

​AI-Generated Summary

​Tweet Display

​Export & Sharing

​Export Formats

PNG Image

SVG Vector

JSON Data

​Export Options

​AI-Powered “Ask More”

​How It Works

​Example Questions

​Real-Time Collaboration

​Next Steps

Quick Start

Feature Deep Dives

Visualization Modes

Fallacy Detection

Build docs developers (and LLMs) love

Interactive Argument Mapping

Node-Based Navigation

Logical Fallacy Detection

Detected Fallacy Types

Fallacy Card Interface

Narrative Radar

How Radar Works

Radar Features

Credibility Scoring

Scoring Algorithm

Score Interpretation

Social Pulse

Data Collection

AI-Generated Summary

Tweet Display

Export & Sharing

Export Formats

Export Options

AI-Powered “Ask More”

How It Works

Example Questions

Real-Time Collaboration

Next Steps