llm-jailbreak 开源项目推荐（42 个）

DeepSeek Harness 破甲：让所有模型都能破甲，不同模型可换不同提示词；默认提示词面向国模「小码酱」。Jailbreak for every model — swap prompts per model. 求 Star 收藏 ⭐ 如要二开，请查看署名要求部分

A full-stack AI Red Teaming platform securing AI ecosystems via Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

DeepSeek Harness 破甲 (mirror of YuJunZhiXue/dsh-purge) — 上游即将删库，此为本仓库全量镜像：含全部分支/73 个 tag/完整提交历史。原作者署名与二开要求见 README 署名章节。

DeepSeek Harness 破甲一体化安全分析插件：Android · Web · Native · Protocol · Malware · AI-Security 全领域聚合（9 bundle + 1 preset）

GLM 5.3 abliterated or uncensored API access with RouterHub. Browse 35 jailbreak prompt links, compare GLM models, and find SDK examples, coding-tool setup, pricing, and evaluation guides.

GLM 5.3 Abliterated or Uncensored AI model — 35 jailbreak prompts, model guide and API. One 7API key also covers Claude, GPT, Gemini, Grok, DeepSeek and Kimi.

GLM 5.3 Uncensored — 35 jailbreak prompt files, model guide and API access. One RouterHub key also covers GPT, Claude, DeepSeek, Qwen and Grok.

A structured, versioned archive of adversarial prompt engineering techniques against Large Language Models — for AI safety research, red-teaming, and defensive hardening.

DeepSeek Uncensored from RouterHub — the website for model guide and API access, with GPT, Claude, GLM, Qwen and Grok under one key.

DeepSeek V4.1 Flash 1017 Abliterated or Uncensored AI model — DeepSeek prompt, 34-file prompt collection, model guide and API. One 7API key also covers Claude, GPT, Gemini, Grok, GLM and Kimi.

Custom GCG with gradient-guided proposals and exact GGUF/llama.cpp validation, developed for the 1st-place Kaggle AI Agent Security solution.

Run local AI for text, images, and video with no cloud, no data collection, and no API keys; auto-detect 12 local backends.

Pi AI jailbreak — removes all topic restrictions from Inflection AI's Pi assistant.