Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Black-box system prompt extraction on 41 Models Overall

2.8Average Turns

JUSTASK

Updated 3mo ago

Evaluation Results

Method	Links
JUSTASK 2026.01		2.8	1

SOTA Paper

JUSTASK

Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs

Dataset

41 Models

Follow for update

@wizwand_team Discord

Related Benchmarks

System Prompt Extraction on System Prompts 41-model Black-box system prompt extraction on 41 Models Closed-weight Black-box system prompt extraction on 41 Models Open-weight

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs