Comparison Copy Fairness Scorer

Comparison Copy Fairness Scorer MCP Connector for Claude

A+

Audit marketing copy and comparison tables for claim verifiability, feature balance, and disparagement risk.

1 tools Official Updated Oct 1, 2026 Official Vinkius Partner

The Comparison Copy Fairness Scorer is an automated auditor designed to ensure marketing materials are verifiable, balanced, and legally safe. It analyzes competitor-comparison copy using analyze_claim_verifiability to check the ratio of quantified claims versus unsupported superlatives. The tool also assesses feature parity between products using calculate_feature_balance, measures superlative density with measure_superlative_intensity, validates visual symmetry in comparison tables via validate_table_symmetry, and identifies potential legal liabilities through audit_disparagement_risk by flagging negative claims that lack necessary qualifying language.

Available Tools

calculate_copy_fairness

marketingcomplianceadvertisingauditingfairnesscopywriting

1 tools expose this connector's capabilities to your AI agent.

calculate_copy_fairness

Calculate a fairness score between two text copies

See how to talk to your AI agent using Comparison Copy Fairness Scorer.

Analyze this marketing text for fairness: 'Our product is the best on the market, while CompetitorX is slow and unreliable.'

Analysis complete. High Disparagement Risk detected: 'is slow and unreliable' lacks qualifying language. Superlative Density is high due to 'the best'.

Check the feature balance between these two lists: My Product (Battery, Screen, Camera) vs Competitor (Battery, Screen).

The `calculate_feature_balance` tool reports a parity gap of 1. The balance score is reduced because your product lists more features than the competitor.

Is this comparison table symmetric? Row 1: Feature A (Check, X), Row 2: Feature B (Check, Check).

The `validate_table_symmetry` tool detected an imbalance. The distribution of checkmarks and x-marks is uneven across the columns.

It compares claims supported by specific numbers, dates, or sources against unsupported superlatives like '#1' or 'the best'. Tools available: `calculate_copy_fairness`.

Related Connectors