{"id":49694690,"url":"https://github.com/ProjectRecon/awesome-ai-agents-security","last_synced_at":"2026-05-23T08:00:46.870Z","repository":{"id":327980487,"uuid":"1112142779","full_name":"ProjectRecon/awesome-ai-agents-security","owner":"ProjectRecon","description":"A living map of the AI agent security ecosystem.","archived":false,"fork":false,"pushed_at":"2025-12-10T15:48:10.000Z","size":34,"stargazers_count":28,"open_issues_count":26,"forks_count":33,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-05-16T01:45:23.108Z","etag":null,"topics":["ai-agents","ai-security","autonomous-agents","awesome","awesome-list","cybersecurity","generative-ai","guardrails","jailbreak","large-language-models","llm","llm-ops","llm-security","prompt-injection","red-teaming","runtime-protection","sandboxing","secure-ai","static-analysis","vulnerability-scanners"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ProjectRecon.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-08T07:58:57.000Z","updated_at":"2026-05-13T15:56:45.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/ProjectRecon/awesome-ai-agents-security","commit_stats":null,"previous_names":["projectrecon/awesome-ai-agent-security"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ProjectRecon/awesome-ai-agents-security","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ProjectRecon%2Fawesome-ai-agents-security","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ProjectRecon%2Fawesome-ai-agents-security/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ProjectRecon%2Fawesome-ai-agents-security/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ProjectRecon%2Fawesome-ai-agents-security/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ProjectRecon","download_url":"https://codeload.github.com/ProjectRecon/awesome-ai-agents-security/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ProjectRecon%2Fawesome-ai-agents-security/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33387656,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-23T04:15:53.637Z","status":"ssl_error","status_checked_at":"2026-05-23T04:15:53.242Z","response_time":53,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agents","ai-security","autonomous-agents","awesome","awesome-list","cybersecurity","generative-ai","guardrails","jailbreak","large-language-models","llm","llm-ops","llm-security","prompt-injection","red-teaming","runtime-protection","sandboxing","secure-ai","static-analysis","vulnerability-scanners"],"created_at":"2026-05-07T23:00:22.200Z","updated_at":"2026-05-23T08:00:46.856Z","avatar_url":"https://github.com/ProjectRecon.png","language":null,"funding_links":[],"categories":["🤖 Agent Orchestration and Loop Safety","Others"],"sub_categories":["Claude Code Specific"],"readme":"# Awesome AI Agents Security \n[![Awesome](https://awesome.re/badge.svg)](https://awesome.re)\n[![License: CC0-1.0](https://img.shields.io/badge/License-CC0_1.0-lightgrey.svg)](http://creativecommons.org/publicdomain/zero/1.0/)\n[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/ProjectRecon/awesome-ai-agents-security/blob/main/CONTRIBUTING.md)\n[![Maintenance](https://img.shields.io/badge/Maintained%3F-yes-green.svg)](https://github.com/ProjectRecon/awesome-ai-agents-security/graphs/commit-activity)\n\nA curated list of open-source tools, frameworks, and resources for securing autonomous AI agents.\n\nThis list is organized by the **security lifecycle** of an autonomous agent, covering red teaming, runtime protection, sandboxing, and governance.\n\n## 📖 Table of Contents\n\n- [Agent Firewalls \u0026 Gateways (Runtime Protection)](#-agent-firewalls--gateways-runtime-protection)\n- [Red Teaming \u0026 Vulnerability Scanners](#-red-teaming--vulnerability-scanners)\n- [Static Analysis \u0026 Linters](#-static-analysis--linters)\n- [Sandboxing \u0026 Isolation Environments](#-sandboxing--isolation-environments)\n- [Guardrails \u0026 Compliance](#-guardrails--compliance)\n- [Benchmarks \u0026 Datasets](#-benchmarks--datasets)\n- [Identity \u0026 Authentication](#-identity--authentication)\n- [Contributing](#-contributing)\n\n---\n\n## 🛡️ Agent Firewalls \u0026 Gateways (Runtime Protection)\n*Tools that sit between the agent and the world to filter traffic, prevent unauthorized tool access, and block prompt injections.*\n\n- **[AgentGateway](https://github.com/agentgateway/agentgateway)** - A Linux Foundation project providing an AI-native proxy for secure connectivity (A2A \u0026 MCP protocols). It adds RBAC, observability, and policy enforcement to agent-tool interactions.\n- **[Envoy AI Gateway](https://gateway.envoyproxy.io/)** - An Envoy-based gateway that manages request traffic to GenAI services, providing a control point for rate limiting and policy enforcement.\n\n## ⚔️ Red Teaming \u0026 Vulnerability Scanners\n*Offensive tools to test agents for security flaws, loop conditions, and unauthorized actions.*\n\n- **[Strix](https://github.com/usestrix/strix)** - An autonomous AI agent designed for penetration testing. It runs inside a docker sandbox to actively probe applications and generate verified exploit capabilities.\n- **[PyRIT](https://github.com/Azure/PyRIT)** - Microsoft’s open-source red teaming framework for generative AI. It automates multi-turn adversarial attacks to test if an agent can be coerced into harmful behavior.\n- **[Agentic Security](https://github.com/msoedov/agentic_security)** - A dedicated vulnerability scanner for agent workflows and LLMs capable of running multi-step jailbreaks and fuzzing attacks against agent logic.\n- **[Garak](https://github.com/leondz/garak)** - The \"Nmap for LLMs.\" A vulnerability scanner that probes models for hallucination, data leakage, and prompt injection susceptibilities.\n- **[A2A Scanner](https://github.com/cisco-ai-defense/a2a-scanner)** - A scanner by Cisco designed to inspect \"Agent-to-Agent\" communication protocols for threats, validating agent identities and ensuring compliance with communication specs.\n- **[Cybersecurity AI (CAI)](https://github.com/aliasrobotics/cai)** - A framework for building specialized security agents for offensive and defensive operations, often used in CTF (Capture The Flag) scenarios.\n\n## 🔍 Static Analysis \u0026 Linters\n*Tools to analyze agent configuration and logic code before deployment.*\n\n- **[Agentic Radar](https://github.com/splx-ai/agentic-radar)** - A static analysis tool that visualizes agent workflows (LangGraph, CrewAI, AutoGen). It detects risky tool usage, permission loops, and maps them to known vulnerabilities.\n- **[Agent Bound](https://github.com/ElPaisano/agent-bound)** - A design-time analysis tool that calculates \"Agentic Entropy\"—a metric to quantify the unpredictability and risk of infinite loops or unconstrained actions in agent architectures.\n- **[Checkov](https://github.com/bridgecrewio/checkov)** - While primarily for IaC, Checkov includes policies for scanning AI infrastructure and configurations to prevent misconfigurations in deployment.\n\n## 📦 Sandboxing \u0026 Isolation Environments\n*Secure runtimes to prevent agents from damaging the host system during code execution.*\n\n- **[SandboxAI](https://github.com/substratusai/sandboxai)** - An open-source runtime for executing AI-generated code (Python/Shell) in isolated containers with granular permission controls.\n- **[Kubernetes Agent Sandbox](https://github.com/kubernetes-sigs/agent-sandbox)** - A Kubernetes Native project providing a Sandbox Custom Resource Definition (CRD) to manage isolated, stateful workloads for AI agents.\n- **[Agent-Infra Sandbox](https://github.com/agent-infra/sandbox)** - An \"All-In-One\" sandbox combining Browser, Shell, VSCode, and File System access in a single Docker container, optimized for agentic tasks.\n- **[OpenHands](https://github.com/All-Hands-AI/OpenHands)** - Formerly OpenDevin, this platform includes a secure runtime environment for autonomous coding agents to operate without accessing the host machine's sensitive files.\n\n## 🚧 Guardrails \u0026 Compliance\n*Middleware to enforce business logic and safety policies on inputs and outputs.*\n\n- **[NeMo Guardrails](https://github.com/NVIDIA/NeMo-Guardrails)** - NVIDIA’s toolkit for adding programmable rails to LLM-based apps. It ensures agents stay on topic, avoid jailbreaks, and adhere to defined safety policies.\n- **[Guardrails](https://github.com/guardrails-ai/guardrails)** - A Python framework for validating LLM outputs against structural and semantic rules (e.g., \"must return valid JSON,\" \"must not contain PII\").\n- **[LiteLLM Guardrails](https://github.com/BerriAI/litellm)** - While known for model proxying, LiteLLM includes built-in guardrail features to filter requests and responses across multiple LLM providers.\n\n## 📊 Benchmarks \u0026 Datasets\n*Resources to evaluate agent security performance.*\n\n- **[CVE Bench](https://github.com/uiuc-kang-lab/cve-bench)** - A benchmark for evaluating an AI agent's ability to exploit real-world web application vulnerabilities (useful for testing defensive agents).\n\n## 🆔 Identity \u0026 Authentication\n*Tools to manage agent identity (non-human identities).*\n\n- **[WSO2](https://github.com/wso2)** - An identity management solution that treats AI agents as first-class identities, enabling secure authentication and authorization for agent actions.\n\n---\n\n## 🤝 Contributing\n\nContributions are welcome! Please read the contribution guidelines first.\n\n1. Fork the project.\n2. Create your feature branch (`git checkout -b feature/AmazingFeature`).\n3. Commit your changes (`git commit -m 'Add some AmazingFeature'`).\n4. Push to the branch (`git push origin feature/AmazingFeature`).\n5. Open a Pull Request.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FProjectRecon%2Fawesome-ai-agents-security","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FProjectRecon%2Fawesome-ai-agents-security","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FProjectRecon%2Fawesome-ai-agents-security/lists"}