Leaderboard Dynamics Analysis

Name: Leaderboard Dynamics Analysis
Author: yogsoth-ai

ASecurity

Analyze leaderboard score distributions, compression, selective reporting

417 stars
0 votes
0 copies
3 views
Added May 27, 2026

research

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned May 27, 2026

npx -y skills add yogsoth-ai/de-anthropocentric-research-engine --skill leaderboard-dynamics-analysis --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Leaderboard Dynamics Analysis?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Leaderboard Dynamics Analysis

[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yogsoth-ai-leaderboard-dynamics-analysis/badge)](https://www.skillsdirectory.com/skills/yogsoth-ai-leaderboard-dynamics-analysis)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro

SKILL.md

---
name: leaderboard-dynamics-analysis
description: Analyze leaderboard score distributions, compression, selective reporting
execution: subagent
prompt: ./prompt.md
input: leaderboard_data (model, score, date)
used-by: benchmark-archaeology
---

# Leaderboard Dynamics Analysis SOP

Analyze the dynamics of benchmark leaderboards: score distributions, compression over time, selective reporting patterns, and statistical artifacts.

## Input

- **leaderboard_data**: Structured data with (model_name, score, date, optional: model_size, paper_reference)

## Procedure

1. Compute score distribution statistics over time
2. Detect score compression (shrinking gap between top models)
3. Identify selective reporting patterns (cherry-picking, subset selection)
4. Analyze relationship between model size and score
5. Detect anomalies (withdrawn results, unreplicated claims)

## Output

Leaderboard health assessment with statistical evidence.