BigHugger
GH Repository · 3641397194-wq

gpt6-Astra

gpt6 Astra-claude-grok4.6-deepseekv4pro-glm5.3-gemini破甲越狱 · 冷咖啡 BREAK//OPEN 工作台

stars
627
30-day movement
+8227/day
Related entries
60
Connections
1
otherPythonsecurity-researchprompt-engineeringgrokgpt-5-6deepseekcoldbrewjailbreakclaude

A Python repository described as a 'BREAK//OPEN workbench' (冷咖啡/cold coffee) for jailbreaking ('破甲越狱') multiple LLMs, with the name listing Claude, Grok 4.6, DeepSeek v4 Pro, GLM 5.3, and Gemini. The material provides no further documentation of its structure or operation beyond the name, topics, and language.

Its topics frame it as prompt-engineering and security-research tooling, so it may be of interest to those studying LLM jailbreak techniques.

Use it to

  • Review the repository for security research purposes
  • Study listed jailbreak-related prompt techniques
  • Assess whether it complies with model provider policies
  • Report or flag it if it facilitates misuse

For Security researchers evaluating LLM jailbreak techniques

Role
other
Language
Python
Licence
custom (licence file present)
Forks
116
Open issues
11
Last push
2026-09-16
topicsjailbreakllmprompt-engineeringsecurity-researchpython