#Safety
Showing 54 of 54 repositories tagged #safety, ranked by stars
👷 TypeScript's largest type utility library
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.
🔐CNCF Security Technical Advisory Group -- secure access, policy control, privacy, auditing, explainability and more!
A reading list for large models safety, security, and privacy (including Awesome LLM Security, Safety, etc.).
Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback
Awesome-Jailbreak-on-LLMs is a collection of state-of-the-art, novel, exciting jailbreak methods on LLMs. It contains papers, codes, datasets, evaluations, and analyses.
保护你的浏览器指纹 | Protect Your Browser Fingerprints | Chrome, Edge, Firefox | 扩展 / Extension
☕ Latte: the safest & truly intuitive templates for PHP. Engine for those who want the most secure PHP sites.
Chinese safety prompts for evaluating and improving the safety of LLMs. 中文安全prompts，用于评估和提升大模型的安全性。
The Destructive Command Guard (dcg) is for blocking dangerous git and shell commands from being executed by agents.
ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers (aka The Norton for OpenClaw)
[EMNLP 2024 Demo] MarkLLM: An Open-Source Toolkit for LLM Watermarking
Awesome-LLM-Robustness: a curated list of Uncertainty, Reliability and Robustness in Large Language Models
Performing security tests inside your CI
⚒️ Changes JsonValues to your custom typescript type.
Bao, a Lightweight Static Partitioning Hypervisor
📝 Generating, validating and processing secure forms in PHP. Handy API, fully customizable, server & client side validation and mature design.
Zebra Crossing: an easy-to-use digital safety checklist
Copy links from the sharing menu with automatic removal of shorteners and trackers.
Opinionated Next JS project boilerplate with TypeScript and Redux
Curated embodied AI list: surveys, VLA models, datasets, simulators, humanoids, robot learning, and safety resources.
Independent research white paper by Jon “GainSec” Gaines examining the security posture of a connected public safety technology ecosystem.
The dispatch doctor prescribes type stability
BeaverTails is a collection of datasets designed to facilitate research on safety alignment in large language models (LLMs).
Code for ACL 2024 paper "TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space"
🔒command line tool checking password safety🔒
在安卓任何聊天软件(例如QQ或微信)中使用加密聊天,基于无障碍服务
[ICML 2024] Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
[ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"
Transparent medical image AI via an image–text foundation model grounded in medical literature
Claude Code plugin: scope-aware protection — allows destructive ops (rm, mv, chmod) within project, blocks them outside. Built for dangerouslySkipPermissions mode.
Datagram Transport Layer Security (DTLS) client.
Audits prompts as behavioral contracts. Pre-write guard for agents that ship code.
Sandboxed Rhai script execution engine with Cedar policy authorization for every system operation.
Swift Codable JSON:API framework
`refined_type` is a library that facilitates type composition, enabling the simple description of efficient validation processes and high runtime safety.
Two-person rule CLI for AI coding agents: peer review and approval required before running potentially destructive commands
MaxVibe - модифицированная версия мессенджера MAX.
macOS command line tool to move files to trash
Template repository with AI agent guardrails, safety protocols, and sprint task framework. For Claude, GPT, Gemini, and all LLMs.
Official implementation of paper: DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers
Open source Android, iOS and Web app for learning about and managing digital and physical security. From how to send a secure message to dealing with a kidnap. Umbrella has best practice guides in over 40 topics in multiple languages. Used daily by people working in high risk countries - journalists, activists, diplomats, business travelers etc.
Security audit Python project dependencies against security advisory databases.
ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark
A fast responsive webapp that allows you to report injured stray animals to the nearest NGOs.
Safe concurrency in D (`shared` made easier)
Open-source toolkit for responsible AI: CLI + SDK to scan code, collect evidence, and generate model cards, risk files, evals, and RAG indexes.
GenoArmory: A Unified Evaluation Framework for Adversarial Attacks on Genomic Foundation Models
This is a python wifi (router) hacker , having ability to search for mikrotic devices around you and get their <MAC> address then extract their user and password
Python agent framework with AgentHarness: planner → generator ⇄ evaluator loops for long-running tasks—plus sessions, tool/safety policies, and Claude, GPT, or Gemini.
Automated Red-Teaming tool for LLM Security. Evaluate system prompts against prompt injection, jailbreak, and data exfiltration attacks.
Check your data which may be stolen every time you visit a site. ⚠️
Anti Virus Tool for windows and linux
One command to make Claude Code safe for autonomous operation. Prevents rm -rf, secret leaks, force-pushes, broken deployments. 890+ example hooks, MIT, zero dependencies — built from real autonomous Claude Code operation and number-for-number real incident reports.