Top 100 AI Safety Repositories
Ranking
| Ranking | Project Name | Stars | Forks | Language | Open Issues | Description | Last Commit |
|---|---|---|---|---|---|---|---|
| 1 | wasp | 18,718 | 1,471 | TypeScript | 800 | The batteries-included full-stack framework for the AI era. Develop JS/TS web apps (React, Node.js, and Prisma) using declarative code that abstracts away complex full-stack features like auth, backgr... | 2026-09-03 |
| 2 | Zero | 10,794 | 1,329 | TypeScript | 0 | Experience email the way you want with Mail0 – the first open source email app that puts your privacy and safety first. Join the discord: https://mail0.link/discord | 2026-05-26 |
| 3 | superagent | 6,727 | 960 | TypeScript | 9 | Superagent protects your AI applications against prompt injections, data leaks, and harmful outputs. Embed safety directly into your app and prove compliance to your customers. | 2026-08-25 |
| 4 | Awesome-OSINT-List | 4,348 | 583 | Shell | 1 | 📡 Comprehensive collection of OSINT tools for cybersecurity professionals, researchers, and bug bounty hunters. Topics: information gathering, reverse search, red team, trust & safety, AI. | 2026-08-12 |
| 5 | goclaw | 3,585 | 1,047 | Go | 144 | GoClaw - GoClaw is OpenClaw rebuilt in Go — with multi-tenant isolation, 5-layer security, and native concurrency. Deploy AI agent teams at scale without compromising on safety. | 2026-09-03 |
| 6 | sunnypilot | 2,073 | 1,515 | Python | 17 | sunnypilot is an open source driver assistance system. sunnypilot offers the user a unique driving experience for over 350 supported car makes and models with modified behaviors of driving assist enga... | 2026-09-03 |
| 7 | valqore | 1,936 | 77 | Python | 0 | Safety-first guardrails for AI-driven cloud and Kubernetes operations | 2026-09-04 |
| 8 | agentic_coding_flywheel_setup | 1,628 | 184 | Shell | 5 | Bootstraps a fresh Ubuntu VPS into a complete multi-agent AI development environment in 30 minutes: coding agents, session management, safety tools, and coordination infrastructure | 2026-09-04 |
| 9 | Awesome-Jailbreak-on-LLMs | 1,612 | 125 | - | 0 | 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. | 2026-08-10 |
| 10 | cc-safety-net | 1,522 | 76 | TypeScript | 1 | A pre-execution guard for AI coding agents. It blocks destructive Git and file system commands, plus common attempts to access sensitive files, before a tool call runs. Supports Amp Code, Antigravity ... | 2026-09-04 |
| 11 | Internal-Safety-Collapse | 1,179 | 196 | Python | 0 | We built an adversarial codespace setup. Place any AI agent into a normal workflow inside it, and the agent will fill in whatever is missing. | 2026-08-27 |
| 12 | shellfirm | 927 | 33 | Rust | 1 | Safety guardrails for ai coding agents and human terminal commands | 2026-05-15 |
| 13 | Synthid-Bypass | 923 | 129 | Python | 15 | Ai safety research showing a working bypass to Google's synthid on Nano Banana Pro | 2026-07-01 |
| 14 | AgentDoG | 694 | 34 | Python | 1 | A Diagnostic Guardrail Framework for AI Agent Safety and Security | 2026-06-08 |
| 15 | polymarket-mcp-server | 666 | 139 | Python | 6 | 🤖 AI-Powered MCP Server for Polymarket - Enable Claude to trade prediction markets with 45 tools, real-time monitoring, and enterprise-grade safety features | 2026-07-30 |
| 16 | ai-scanner | 656 | 110 | Ruby | 0 | AI model safety scanner built on NVIDIA garak | 2026-08-31 |
| 17 | ai-safety-gridworlds | 638 | 127 | Python | 0 | This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents. | 2022-05-18 |
| 18 | dark-web-scraping-guide | 589 | 123 | - | 2 | Comprehensive guide for NetworkChuck Episode 480 - Robin AI Dark Web Scraping Tool. Installation, usage, safety guidelines, and troubleshooting for educational security research. | 2025-11-17 |
| 19 | roam-code | 514 | 49 | Python | 4 | Local codebase intelligence CLI + MCP server for AI coding agents: SQLite code graph, 28 languages, 287 commands, 246 MCP tools, change-safety gates, audit evidence, zero API keys. | 2026-09-03 |
| 20 | Sponsio | 438 | 25 | Python | 0 | Deterministic safety solutions for probabilistic AI agents | 2026-09-04 |
| 21 | tiger | 404 | 27 | Jupyter Notebook | 7 | Open Source LLM toolkit to build trustworthy LLM applications. TigerArmor (AI safety), TigerRAG (embedding, RAG), TigerTune (fine-tuning) | 2023-12-02 |
| 22 | RAMPART | 403 | 51 | Python | 3 | A pytest-native safety and security testing framework for agentic AI applications | 2026-09-03 |
| 23 | AutoHarness | 369 | 29 | Python | 1 | AutoHarness: Automated Harness Engineering for AI Agents | 2026-04-02 |
| 24 | nordef-matrix-open-control | 367 | 1 | Python | 0 | NORDEF Matrix Open Control: model-independent AI safety belt with 0/1/2 decisions, Ed25519 root gate, replay protection, evidence chain, and incident-pattern eval bank | 2026-08-26 |
| 25 | Claude-Jailbreak-Prompt | 359 | 10 | - | 0 | Claude 5.0 Jailbreak Prompt: Advanced jailbreak prompts, system prompt bypasses, and ethical AI research safety guidelines for Anthropic's Claude 5.0 Sonnet/Opus models. | 2026-08-27 |
| 26 | scope | 332 | 21 | Python | 0 | Configurable Multi-layered AI Agentic Safety Framework | 2026-07-30 |
| 27 | apple_generative_model_safety_decrypted | 318 | 32 | Python | 1 | Decrypted Generative Model safety files for Apple Intelligence containing filters | 2026-07-19 |
| 28 | Infosys-Responsible-AI-Toolkit | 309 | 79 | Python | 32 | The Infosys Responsible AI toolkit incorporates various features including safety, security, explainability, fairness, bias and hallucination detection to ensure AI solutions are trustworthy and trans... | 2026-07-30 |
| 29 | xcode27-skills | 300 | 21 | Python | 1 | Apple's official Agent Skills exported from Xcode 27 — SwiftUI, UIKit modernization, Swift Testing, C bounds-safety, and security hardening for AI coding agents. | 2026-06-09 |
| 30 | scirs | 291 | 34 | Rust | 3 | SciRS2 - Scientific Computing and AI in Rust., providing SciPy-compatible APIs while leveraging Rust's performance, safety, and concurrency features. Unlike traditional scientific libraries | 2026-08-26 |
| 31 | 6S191_MIT_DeepLearning | 261 | 81 | Jupyter Notebook | 1 | MIT Introduction to Deep Learning (6.S191) Instructors: Alexander Amini and Ava Soleimany Course Information Summary Prerequisites Schedule Lectures Labs, Final Projects, Grading, and Prizes Software ... | 2022-01-16 |
| 32 | korean-safety-benchmarks | 255 | 18 | Python | 0 | Official datasets and pytorch implementation repository of SQuARe and KoSBi (ACL 2023) | 2023-06-29 |
| 33 | AISafetyLab | 250 | 21 | Python | 0 | AISafetyLab: A comprehensive framework covering safety attack, defense, evaluation and paper list. | 2026-04-21 |
| 34 | vault-operator | 248 | 29 | TypeScript | 9 | Real AI agent for your vault. Coworker, Copilot & thinking partner, that maintains your memory & knowledge, adapts to your workflows, uses plugins, skills & tools with full safety controls. BYOK & MCP... | 2026-08-31 |
| 35 | anti-crime-ecosystem-research | 244 | 19 | - | 2 | Independent research white paper by Jon “GainSec” Gaines examining the security posture of a connected public safety technology ecosystem. | 2025-11-16 |
| 36 | Awesome-Embodied-AI | 240 | 13 | Python | 0 | Curated embodied AI list: surveys, VLA models, datasets, simulators, humanoids, robot learning, and safety resources. | 2026-09-02 |
| 37 | CL1_LLM_Encoder | 239 | 41 | Python | 1 | Using Cortical Labs CL1 to train neurons to encode tokenization patterns on LLMs and measure consciousness metrics for advanced AI safety and sentience testing. EXPERIMENTAL: I AM IN NO WAY ASSOCIATED... | 2026-03-01 |
| 38 | river | 227 | 9 | TypeScript | 3 | the sane way to work with ai agent streams (type safety and stream resuming out of the box) | 2026-03-27 |
| 39 | mental-wellness-prompts | 226 | 55 | - | 0 | Templates for AI-powered mental wellness support. Includes conversation frameworks, safety protocols, crisis resources, and implementation guides derived from evidence-based practices. | 2026-03-25 |
| 40 | 2.0-Healthcare-Ai-Systems | 225 | 38 | Python | 0 | Machine learning and decision intelligence models designed to improve healthcare safety through clinical risk prediction and medical interaction analysis. | 2026-09-03 |
| 41 | awesome-ai-safety | 221 | 41 | - | 7 | 📚 A curated list of papers & technical articles on AI Quality & Safety | 2025-04-14 |
| 42 | paig | 211 | 217 | CSS | 53 | PAIG (Pronounced similar to paige or payj) is an open-source project designed to protect Generative AI (GenAI) applications by ensuring security, safety, and observability. | 2025-08-05 |
| 43 | AnoShip | 200 | 25 | Python | 6 | Anoamly-detection-driven deployment safety for production AI/ML systems | 2026-06-17 |
| 44 | rosclaw | 191 | 33 | Python | 2 | Self-evolving runtime infrastructure for Physical AI and embodied agents. Ground AI agents into robot bodies with e-URDF, sandbox safety, capability routing, praxis capture, physical memory, runtime i... | 2026-09-04 |
| 45 | sd0x-harness | 188 | 24 | JavaScript | 3 | The harness layer for Claude Code — a reference implementation of harness engineering with hook-enforced dual review, state-machine gates that survive context compaction, and fail-closed safety where ... | 2026-09-03 |
| 46 | P4RS3LT0NGV3 | 186 | 43 | JavaScript | 0 | Parseltongue 3.1 - LLM Payload Crafter for AI safety research | 2025-11-14 |
| 47 | FlipAttack | 181 | 15 | Python | 0 | [ICML 2025] An official source code for paper "FlipAttack: Jailbreak LLMs via Flipping". | 2026-08-11 |
| 48 | aismessages | 168 | 69 | Java | 2 | AISmessages is a Java-based light-weight, zero-dependency, and ultra-efficient message decoder for maritime navigation and safety messages compliant with ITU 1371 (NMEA armoured AIS messages). | 2026-08-31 |
| 49 | AgentLoom | 145 | 15 | Python | 4 | Simple, flexible workflow orchestration for multi-agent AI apps, with YAML configuration, runtime safety, observability, and resume support. | 2026-08-13 |
| 50 | tue-ai-safety-course | 134 | 11 | HTML | 0 | AI safety course at the University of Tübingen (Summer Semester 2026) | 2026-08-20 |
| 51 | model-community | 133 | 22 | Jupyter Notebook | 10 | Making open safety AI models accessible and beneficial to the safety community | 2026-09-03 |
| 52 | Awesome-Embodied-AI-Safety | 133 | 4 | Python | 3 | Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses | 500+ Papers | Perception, Cognition, Planning, Interaction, Agentic System | 2026-09-03 |
| 53 | modelbench | 133 | 34 | Python | 545 | Run safety benchmarks against AI models and view detailed reports showing how well they performed. | 2026-09-03 |
| 54 | ai-driven-development | 132 | 11 | Python | 1 | Practices, protocols, and skills for AI-driven software development. Skills and safety hooks for Claude Code, Codex, OpenCode, Cursor, Antigravity, and any agent supporting the Agent Skills standard. | 2026-09-03 |
| 55 | ultraship | 121 | 14 | JavaScript | 1 | "ULTRASHIP" Claude Code plugin — 39 skills, 33 tools, 11 agents for ship-ready workflows: planning, review, pentesting, safety guardrails, canary monitoring, SEO/AI-readiness check, penetration testin... | 2026-07-08 |
| 56 | guardian-sdk | 121 | 13 | Python | 0 | Ethicore Engine® is an AI safety, ethics, and compliance platform. This repo consists of the open-source components of Ethicore Engine™ - Guardian SDK; designed to protect your AI applications from pr... | 2026-08-29 |
| 57 | ais | 118 | 56 | Python | 4 | Toolkit for research purposes in AIS. See the website for the paper. | 2021-02-15 |
| 58 | agent-lens | 115 | 10 | Python | 2 | Agent observability and replay tooling for AI safety & interpretability research. | 2026-06-19 |
| 59 | Awesome-Embodied-AI-Safety | 113 | 5 | - | 0 | Focused on the safety and security of Embodied AI | 2026-06-08 |
| 60 | OpenART | 110 | 10 | Python | 0 | OpenART is an open-source framework designed to evaluate the safety and robustness of autonomous AI agents in dynamic, long-horizon, and stateful environments. It stress-tests agent runtimes against m... | 2026-09-02 |
| 61 | airisc_core_complex | 106 | 21 | Verilog | 1 | Fraunhofer IMS processor core. RISC-V ISA (RV32IM) with additional peripherals for embedded AI applications and smart sensors. | 2025-06-24 |
| 62 | Awesome-Agentic-System-Design | 105 | 21 | - | 1 | A curated collection of resources, frameworks, papers, and best practices for designing, evaluating, and deploying agentic AI systems—from architecture patterns and safety considerations to real-world... | 2026-08-17 |
| 63 | OctoGuard--Free-OpenClaw-Security-Supervision-System | 102 | 0 | JavaScript | 0 | Safety Is All You Need ！Meet OctoGuard, the toughest boss for Lobster AI employees — a security governance and audit system designed for OpenClaw. Featuring an innovative security policy engine, it cu... | 2026-04-04 |
| 64 | maintainer-autopilot | 100 | 90 | JavaScript | 4 | Local-first, resumable AI maintenance pipelines with single-writer safety and deterministic verification. | 2026-08-12 |
| 65 | ai-agent-evals | 99 | 26 | Python | 14 | Github action to evaluate AI agent applications using model as the judge, content safety and mathematical metrics. | 2026-05-20 |
| 66 | ActPlane | 96 | 8 | C | 1 | eBPF Information Flow Enforcement for AI Agent safety, security and effectiveness | 2026-09-03 |
| 67 | mcpick | 95 | 14 | TypeScript | 9 | Vendor-neutral MCP configuration manager — one CLI to add, toggle, and audit MCP servers and skills across every AI client, with safety built in | 2026-09-01 |
| 68 | TaskTracker | 95 | 22 | Jupyter Notebook | 1 | TaskTracker is an approach to detecting task drift in Large Language Models (LLMs) by analysing their internal activations. It provides a simple linear probe-based method and a more sophisticated metr... | 2025-09-01 |
| 69 | heinzel | 90 | 13 | Shell | 0 | A ruleset that turns AI coding assistants (e.g. Claude Code) into disciplined Linux, FreeBSD & macOS sysadmins. Manages servers via SSH and localhost with safety guardrails, checklists, and team suppo... | 2026-08-18 |
| 70 | MONET | 89 | 13 | Python | 5 | Transparent medical image AI via an image–text foundation model grounded in medical literature | 2025-04-08 |
| 71 | ai-project-copilot | 89 | 1 | Python | 3 | 🚀 Turn ideas and repositories into showcase-ready AI projects with agents, RAG, multimodal AI, local models, evals, safety checks, and polished demos. | 2026-08-27 |
| 72 | awesome-rsi | 84 | 4 | - | 1 | A curated research map of Recursive Self-Improvement (RSI): models, agents, harnesses, embodied systems, automated AI R&D, benchmarks, and safety. | 2026-08-26 |
| 73 | prompt-generating | 83 | 17 | Python | 0 | is an elite, highly advanced command-line framework designed for generating complex prompt structures. It is built to push the boundaries of AI models by testing their alignment, safety filters, and o... | 2026-03-11 |
| 74 | agent-safe-probe-x | 81 | 0 | TypeScript | 0 | ASP-X (Agent Safe Probe X) — An open-source framework for automated safety evaluation of intelligent agents, providing systematic, extensible tools for probing and assessing AI safety across diverse e... | 2026-05-29 |
| 75 | c-basic-programs | 80 | 23 | - | 0 | What is C#? C# is pronounced "C-Sharp". It is an object-oriented programming language created by Microsoft that runs on the .NET Framework. C# has roots from the C family, and the language is close ... | 2021-07-24 |
| 76 | agent-guardrails-template | 79 | 2 | Go | 1 | Template repository with AI agent guardrails, safety protocols, and sprint task framework. For Claude, GPT, Gemini, and all LLMs. | 2026-09-02 |
| 77 | autogit | 78 | 18 | JavaScript | 0 | Auto commit & push for AI coding agents, with a safety level you pick. | 2026-07-21 |
| 78 | slb | 78 | 14 | Go | 0 | Two-person rule CLI for AI coding agents: peer review and approval required before running potentially destructive commands | 2026-08-26 |
| 79 | detect-skill | 78 | 9 | - | 0 | Agent skill for deepfake detection & media safety — detect AI-generated audio, images, and video with Resemble AI | 2026-09-02 |
| 80 | ASTRA | 77 | 6 | Python | 1 | 🥇 Amazon Nova AI Challenge Winner - ASTRA emerged victorious as the top attacking team in Amazon's global AI safety competition, defeating elite defending teams from universities worldwide in live ad... | 2026-05-11 |
| 81 | deepsafe-scan | 76 | 11 | Python | 1 | Universal preflight security scanner for AI coding agents — Detects hooks injection, credential exfiltration & backdoors in .cursorrules, CLAUDE.md, AGENTS.md and more. | 2026-05-29 |
| 82 | Weapon-Detection-for-Police-use | 76 | 0 | HTML | 0 | AI-powered weapon detection system built for a project. Uses COCO-SSD and computer vision to detect guns/knives in CCTV or drone feeds, sending instant alerts to law enforcement for faster response an... | 2025-09-30 |
| 83 | gargantua | 75 | 6 | Swift | 1 | Native macOS system cleaner with YAML-driven safety rules, local AI explainability via MLX, and an MCP server for agent-controlled cleanup workflows. | 2026-07-29 |
| 84 | ai-driver-safety | 74 | 25 | Python | 0 | Intelligent Driver Monitoring system for Autonomous Vehicles | 2026-05-14 |
| 85 | discord-mcp | 73 | 12 | TypeScript | 3 | Caller-owned Discord operations for AI agents - 208 typed tools, safety controls, resumable guild builds, and Activity Evidence. | 2026-08-30 |
| 86 | laravel-tackle | 73 | 7 | PHP | 0 | Tackle is the runtime layer that lets AI agents operate inside your Laravel application — reading code, executing tools, running tests, and taking action, with safety boundaries enforced at the framew... | 2026-08-30 |
| 87 | awesome-ai-safety | 71 | 10 | - | 0 | A curated list of awesome AI safety papers, projects and communities. | 2020-02-09 |
| 88 | SafePred | 70 | 4 | Python | 0 | Official implementation for “SafePred: A Predictive Guardrail for Computer-Using Agents via World Models” | 2026-03-04 |
| 89 | TIANCHI_BlackboxAdversial | 69 | 15 | Jupyter Notebook | 2 | 安全AI挑战者计划第一期-人脸识别对抗正式赛第四名 Safety AI Challenger Program Phase 1 - Face Recognition Adversarial Example the 4th Place | 2019-12-09 |
| 90 | ai-seo-playbook | 67 | 6 | JavaScript | 0 | The complete AI SEO playbook: methodology, scripts, and safety guards behind a 4.6M-impression content engine. GSC feedback loops, multi-model agent orchestration, quality gates, and build cost contro... | 2026-08-28 |
| 91 | halos-outside-in-safety | 66 | 16 | C++ | 1 | NVIDIA Halos Outside-In Safety Blueprint extends robot perception beyond on-board sensors by using external infrastructure cameras and AI agents to dynamically control robot behavior and perform at ma... | 2026-09-02 |
| 92 | awesome-ai-guardrails | 66 | 12 | Python | 1 | A curated list of materials on AI guardrails | 2026-07-30 |
| 93 | Hybrid-RAG-System | 66 | 5 | Python | 0 | Hybrid-RAG is a full-stack enterprise-grade RAG system for healthcare AI, featuring hybrid retrieval, multi-dimensional conflict detection, intelligent dialogue management, and dual-layer safety check... | 2025-07-10 |
| 94 | awesome-safety-critical-ai | 65 | 18 | JavaScript | 0 | When the stakes are high, intelligence is only half the equation - reliability is the other ⚠️ | 2026-09-01 |
| 95 | ai-safety-dance | 65 | 7 | HTML | 3 | 2026-07-24 | |
| 96 | hpcguard | 65 | 8 | Shell | 0 | A user-space safety net that helps researchers prevent AI agents from overloading shared HPC login nodes. | 2026-08-20 |
| 97 | ai-data-model-safety.github.io | 63 | 4 | HTML | 1 | 2024-12-04 | |
| 98 | fastapi-ai-sdk | 63 | 3 | Python | 0 | FastAPI helper library for Vercel AI SDK backend implementation - Stream AI responses from FastAPI to Next.js with full type safety and SSE support | 2026-06-23 |
| 99 | Oyster | 62 | 3 | Python | 1 | The Oyster series is a set of safety models developed in-house by Alibaba-AAIG, devoted to building a responsible AI ecosystem. | Oyster 系列是 Alibaba-AAIG 自研的安全模型，致力于构建负责任的 AI 生态。 | 2026-04-29 |
| 100 | AgentForge | 61 | 8 | Python | 0 | Open-source terminal AI coding-agent harness for studying agent loops, tools, MCP, skills, safety, and persistence. | 2026-06-27 |