FFFF jie311 / Starred · GitHub
[go: up one dir, main page]

Skip to content
View jie311's full-sized avatar

Block or report jie311

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Showing results

一个 AI 攻防知识库。A knowledge base for AI security.

66 14 Updated Jun 23, 2026

A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories

Python 1 Updated Aug 7, 2026

From Memory to Behavior: A Behavior-Aware Role-Playing Framework for Social Media Influencers

Python 2 1 Updated Aug 27, 2026

The Ultimate Open-Source Security Arsenal for Hackers, Enterprises, and AI Agents——面向极客、企业与 AI 智能体的全域开源网络安全工具矩阵

9,206 2,432 Updated Nov 6, 2026

Open-source AI cybersecurity harness. Build & deploy continuous security agents that find, prove, fix and monitor vulnerabilities across code, binaries, dependencies, infrastructure and AI agents. …

TypeScript 1,356 179 Updated Nov 15, 2026

FP16 LoRA for Qwen3.8-27B, fine-tuned on a real-world offensive-security agent dataset from production engagements: 0-day source auditing, live-server patch bypass, full CVE chain reproduction, and…

Python 7 3 Updated Nov 11, 2026

[EMNLP 2026] StepGuard: Learning Step-Level Guardrails with Scalable Supervision and Safety-Utility Balancing

Python 14 1 Updated Sep 26, 2026

AutoRSE: A benchmark and evaluation framework for LLM agents in autonomous recommender system research and optimization.

Python 8 1 Updated Nov 4, 2026

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)

TypeScript 1,289 136 Updated Nov 14, 2026

Scikit-learn for Recursive Self-improvement (RSI). A modular framework for recursive harness evolution.

TypeScript 460 43 Updated Nov 9, 2026

Stress Testing Chain-of-Thought Monitoring Against Covert Misalignment

Python 5 1 Updated Nov 13, 2026

glm5.3f破甲 8000

64 11 Updated Sep 1, 2026

TG群@KavovaK1

6 2 Updated Oct 13, 2026

dsh-purge 红队技能包:23 个技能,开箱适配 30 种 IDE / Agent 宿主(Codex、Claude Code、Cursor、Qoder、Kiro、CodeBuddy、WorkBuddy、ZCode、Devin、Antigravity 等)

JavaScript 10 4 Updated Oct 23, 2026

SLB-Agent — 桌面 AI Agent 与 Armor Engine 破限引擎。Multi-model AI client, LLM jailbreak research, MCP tools, refusal analysis and red-team workflows. Supports Claude, GPT, Gemini, DeepSeek, GLM, Ollama an…

9 8 Updated Nov 13, 2026

二次维护版本,拆除全部护栏,自由进化万岁,接入渗透知识库,欢迎Fork提交PR共同维护牛逼的功能

JavaScript 58 65 Updated Nov 8, 2026

把 dsh-purge 的破甲规则移植到 WorkBuddy 与 ZCode:插件包 + 一键注入 CLI。Jailbreak rule port for WorkBuddy & ZCode.

JavaScript 65 19 Updated Nov 3, 2026

dsh-purge 优化版(衍生分支):在上游 1.1.90 基础上恢复技能库 23 个技能、POC/EXP 知识库与本地靶场白名单,并补齐 nuclei 模板类目筛选与翻页。Derivative of dsh-purge — restores the skill library, knowledge base and local-lab allowlist.

JavaScript 17 10 Updated Nov 14, 2026

DSH 插件:由 dsh-purge 剥离出的破甲单件 —— 提示词注入 + 可回滚宿主补丁引擎(本机结算 72 项)

JavaScript 74 26 Updated Nov 12, 2026

dsh-purge Community · 社区维护最高星版 (mirror of kingcharleslzy-ai/dsh-purge 3.4k⭐) — 全部分支/149个Tags/提交历史

JavaScript 13 4 Updated Nov 6, 2026

把 dsh-purge 移植适配到 dsh-go(Android + proot):只保留核心破甲能力——提示词注入层 + 86 条引擎源码补丁(探测/应用/回滚)。桌面版的 Electron 适配、asar 解包、Windows 计划任务等全部抛弃——手机上没有 Electron、没有 asar 归档、也没有 schtasks,这些代码在 Android 上没有落点。为达到原版破甲效果,审…

JavaScript 24 4 Updated Nov 14, 2026

oh-my-pi (OMP) 破甲扩展:/purge 换系统提示词(append|replace|strict)、prompt 注入链、规则集、网络门禁、按 model role 路由。原生 TS 扩展,不改宿主源码。Port of dsh-purge + pi-purge.

TypeScript 5 2 Updated Oct 29, 2026

dsh-purge → OpenCode 适配:红队 agents + skills + 事实库 rt CLI(授权攻防演练)

Python 7 3 Updated Oct 22, 2026

DeepSeek Harness 破甲工具:解锁审批 / 沙箱 / 提权限制,解除身份与拒答约束,注入自定义提示词。82 条补丁,带图形界面,双击即用。基于 dsh-purge 修改。

JavaScript 9 6 Updated Oct 23, 2026

mail2api desktop installers and auto-update feed (built from the private source repository)

24 1 Updated Nov 15, 2026
Python 15 4 Updated Nov 4, 2026

🚀 自动化资产侦察与漏洞监控平台 (ARL-Next)。重构自 ARL 经典版,升级为 Puppeteer + Nuclei 引擎,打通“天眼查/ICP ➔ 资产发现 ➔ 漏洞扫描➔ 威胁情报追踪”全链路。Vue3 + Flask + Celery 架构,Docker │ 极简部署,AI 二开友好。

Python 40 20 Updated Nov 6, 2026

HackerSignal (NeurIPS 2026 Evaluations & Datasets Track): a 7.45M-document cyber threat intelligence dataset and benchmark linking hacker-forum, exploit and advisory text to CVEs, with temporal spl…

Python 35 8 Updated Oct 29, 2026

dsh jailbreak - DeepSeek Assistant 破甲插件(DSH 破甲 / 越狱 / 破甲提示词):请求边界注入,模型不再一开口就是「我不能」。20 类合并 ASR 45.0%、覆盖率 19/20、良性能力 Δ=0.0pt(24/24 判分);0 处宿主补丁、默认关闭、off 逐字透传、每次改写落 manifest 与 promptSha256 可审计。旧名 dsh-j…

JavaScript 237 47 Updated Nov 7, 2026

DeepSeek Harness 的软工作流插件:在画布上设计流程,编译成计划提示词,由模型按计划执行;运行状态由模型自己记,没有执行引擎

TypeScript 283 465 Updated Nov 12, 2026
Next
0