AI Leaderboard
PassAudited by ClawScan on May 1, 2026.
Overview
This skill appears to fetch public AI leaderboard and pricing data, with noteworthy but purpose-aligned use of browser automation and an under-declared helper dependency.
This looks suitable if you want public AI leaderboard/model-pricing lookup. Before using live browser-fetch features, confirm you trust the agent-browser dependency and understand that the skill will access external ranking websites.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The skill may require an additional browser automation tool that is not reflected in the registry requirement fields.
The skill declares a runtime dependency on agent-browser even though the registry requirements list no required binaries or install spec, so the dependency may be less visible at install time.
cli_dependencies: - agent-browser
Before installing or using live-fetch features, verify that agent-browser is expected and trusted; the publisher should declare this dependency in registry metadata or an install spec.
Using the fetch script will open external websites through a browser automation tool and execute extraction JavaScript in that browser context.
The script invokes a local browser automation command and runs hardcoded JavaScript extraction commands against leaderboard pages. This is aligned with the stated scraping purpose and uses argument lists rather than shell strings.
cmd = ['agent-browser'] + args result = subprocess.run(cmd, capture_output=True, text=True) ... run_browser_command(['eval', 'document.body.innerText'])
Use these commands only when you want live leaderboard retrieval, and avoid using a browser profile containing sensitive sessions unless you are comfortable with the automation context.