awesome-repositories.com
المدونة
awesome-repositories.com

اكتشف أفضل مستودعات المصادر المفتوحة باستخدام بحث مدعوم بالذكاء الاصطناعي.

استكشفعمليات بحث منسقةبدائل مفتوحة المصدربرمجيات ذاتية الاستضافةالمدونةخريطة الموقع
المشروعحولكيفية ترتيب النتائجالصحافةخادم MCP
قانونيالخصوصيةالشروط
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·

18 مستودعات

Awesome GitHub RepositoriesAgent Security

Frameworks and auditing tools designed to secure agent configurations and prevent vulnerabilities in automated systems.

Distinguishing note: Focuses on security auditing specifically for agentic architectures rather than general application security.

Explore 18 awesome GitHub repositories matching security & cryptography · Agent Security. Refine with filters or upvote what's useful.

Awesome Agent Security GitHub Repositories

اعثر على أفضل المستودعات باستخدام الذكاء الاصطناعي.سنبحث عن أفضل المستودعات المطابقة باستخدام الذكاء الاصطناعي.
  • affaan-m/eccالصورة الرمزية لـ affaan-m

    affaan-m/ECC

    221,981عرض على GitHub↗

    ECC هو إطار عمل لأوركسترا وكلاء LLM ومجموعة أدوات ذكاء اصطناعي عبر المنصات مصممة لتنسيق سير العمل متعدد النماذج. يوفر نظاماً لإدارة أدوار الوكلاء المتخصصة، والمهارات القابلة لإعادة الاستخدام، والتخطيط المهيكل لتنفيذ مهام تطوير البرمجيات المعقدة عبر محررات كود مختلفة مدعومة بالذكاء الاصطناعي. يتميز المشروع كمدير لبروتوكول سياق النموذج (Model Context Protocol)، حيث يوفر طبقة تكوين لدمج الخوادم الخارجية وتدقيق تنفيذ الأدوات. كما ينفذ بيئة أمنية للوكلاء (agentic security sandbox) تقيد الوصول إلى الملفات الحساسة وتفحص تسرب الأسرار لتأمين سير العمل المستقل. يغطي إطار العمل مجالات قدرة واسعة بما في ذلك أتمتة سير عمل البرمجة بالذكاء الاصطناعي مع حواجز حماية التطوير القائم على الاختبار، وتحسين تكلفة النموذج من خلال التوجيه الذكي، وإدارة الذاكرة المعزولة الحالة. كما يتضمن أدوات لفرض معايير البرمجة الخاصة باللغة وإدارة سلوكيات الوكلاء عبر بيئات تطوير متكاملة مختلفة. تتم إدارة النظام من خلال واجهة سطر أوامر تتعامل مع تثبيت الأدوات، وإصلاح التكوين، ونشر إعدادات الأدوات المسبقة.

    Provides a security layer that scans for secret leakage and restricts file access in autonomous workflows.

    JavaScript
    عرض على GitHub↗221,981
  • affaan-m/everything-claude-codeالصورة الرمزية لـ affaan-m

    affaan-m/everything-claude-code

    216,499عرض على GitHub↗

    Everything Claude Code هو إطار عمل وكيل مصمم لتنسيق سير عمل تطوير البرمجيات المعقدة من خلال تفويض الوكلاء الفرعيين المتخصصين. يعمل كطائرة تحكم تدير سلوك الوكيل، والوصول إلى الأدوات، وكفاءة نافذة السياق، مما يسمح للمطورين بتقسيم المهام الكبيرة إلى عمليات فرعية مركزة ومحددة النطاق تمنع تحميل النظام الزائد. يتميز إطار العمل بطبقة أمن وأتمتة قوية تتضمن التحليل الثابت الآلي واختبار الاختراق العدائي لتدقيق تكوينات الوكيل. وهو يتيح إنشاء أنماط سلوكية قابلة لإعادة الاستخدام وتسلسلات أتمتة، والتي يمكن مشاركتها عبر البيئات كمهارات معيارية. من خلال مزامنة التكوينات الخاصة بالمشروع وتعليمات الدردشة، فإنه يضمن بقاء معايير البرمجة وقيود الأمان متسقة عبر كل من سطر الأوامر وبيئات التطوير المتكاملة. بعيداً عن قدرات الأوركسترا الأساسية، يوفر المشروع أدوات شاملة لإدارة التكاليف التشغيلية أثناء الجلسات طويلة الأمد. يتضمن آليات لتحسين الرموز الديناميكي، وإدارة حالة الجلسة، والخطافات القائمة على الأحداث التي تطلق نصوص التحقق أو فرض الجودة. يدعم النظام أيضاً استخراج الأنماط المتكررة من تاريخ التحكم في الإصدار لتوليد مجموعات مهارات متخصصة، مما يزيد من تبسيط مهام التطوير المتكررة.

    A set of automated tools and validation hooks designed to audit agent configurations and prevent vulnerabilities during development.

    JavaScriptai-agentsanthropicclaude
    عرض على GitHub↗216,499
  • github/awesome-copilotالصورة الرمزية لـ github

    github/awesome-copilot

    35,119عرض على GitHub↗

    Awesome Copilot is a comprehensive framework for autonomous software development, providing the infrastructure to orchestrate multi-agent teams and automate complex coding workflows. It functions as a centralized platform for managing AI-driven development, enabling developers to deploy specialized agents that interact with local files, terminal commands, and external APIs to execute end-to-end software delivery tasks. The project distinguishes itself through its focus on governance and extensibility, offering a suite of security controls, policy-based execution guardrails, and audit trails t

    Checks agent-based systems against industry security standards to identify vulnerabilities and ensure compliance.

    Pythonaigithub-copilothacktoberfest
    عرض على GitHub↗35,119
  • huggingface/smolagentsالصورة الرمزية لـ huggingface

    huggingface/smolagents

    27,885عرض على GitHub↗

    This framework provides a development toolkit for building autonomous agents that utilize language models to solve complex, non-deterministic tasks. Its core design centers on a code-executing architecture where agents generate and run Python code snippets to perform logic, data manipulation, and tool interactions. By moving beyond structured data formats, the system enables agents to manage program flow and object state through iterative reasoning cycles. The project distinguishes itself through its focus on code-based agent implementation and secure execution environments. Developers can ch

    Implements security measures and sandboxing to protect against risks associated with agent-generated code.

    Python
    عرض على GitHub↗27,885
  • fosrl/pangolinالصورة الرمزية لـ fosrl

    fosrl/pangolin

    21,255عرض على GitHub↗

    Pangolin is a zero-trust remote access platform designed to provide secure, identity-aware connectivity to private network resources. It functions as a cloud-native network controller that orchestrates encrypted tunnels, traffic routing, and access policies across distributed environments. By leveraging WireGuard for secure data transport, the platform enables authenticated access to internal web applications, terminal sessions, and remote desktops without exposing services to the public internet. The platform distinguishes itself through a declarative infrastructure model that synchronizes n

    Enforces token-based authentication for secure data exchange between monitoring agents and the central dashboard.

    TypeScriptcrowdsecdockerhome-lab
    عرض على GitHub↗21,255
  • nirdiamant/agents-towards-productionالصورة الرمزية لـ NirDiamant

    NirDiamant/agents-towards-production

    17,375عرض على GitHub↗

    This project is a comprehensive framework for developing, orchestrating, and deploying autonomous agents. It provides a structured environment for building agents that utilize reasoning loops to perform multi-step tasks, manage state through graph-based workflows, and interact with external tools. By mapping unstructured model outputs into typed schemas, the framework ensures reliable integration with downstream application logic. The platform distinguishes itself through a focus on production-grade reliability and security. It incorporates hybrid memory systems that combine vector embeddings

    Implements safety protocols, input validation, and human-in-the-loop workflows to secure agent operations.

    Jupyter Notebookagentagent-frameworkagents
    عرض على GitHub↗17,375
  • aws/aws-cdkالصورة الرمزية لـ aws

    aws/aws-cdk

    12,817عرض على GitHub↗

    The AWS Cloud Development Kit is an infrastructure-as-code framework that enables developers to define and provision cloud resources using familiar programming languages. By utilizing construct-based synthesis, it translates high-level, object-oriented code into declarative templates, allowing for the automated management of complex cloud environments through a centralized, code-driven control plane. The framework distinguishes itself through its ability to model infrastructure as a dependency-aware resource graph, ensuring that components are provisioned and updated in the correct order. It

    Manages authentication and access control for agents by integrating with existing identity providers.

    TypeScriptawscloud-infrastructurehacktoberfest
    عرض على GitHub↗12,817
  • crowdsecurity/crowdsecالصورة الرمزية لـ crowdsecurity

    crowdsecurity/crowdsec

    12,574عرض على GitHub↗

    CrowdSec is a collaborative, distributed security engine designed for threat detection and infrastructure protection. It functions as an intrusion detection system that parses logs and network traffic to identify malicious patterns, utilizing a bucket-based threshold detection model to aggregate events and trigger alerts. The platform is built on a modular architecture that includes a centralized local API server for managing security signals and a relational database for persistent storage of remediation decisions. What distinguishes the project is its decoupled enforcement model, which offl

    Allows administrators to verify the connectivity and operational health of security agents.

    Goattacks-preventiondetectionids
    عرض على GitHub↗12,574
  • golang-jwt/jwtالصورة الرمزية لـ golang-jwt

    golang-jwt/jwt

    8,908عرض على GitHub↗

    This project is a JSON Web Token implementation library and toolkit for encoding, signing, and validating tokens. It provides the necessary functions to manage token claims and payloads for user authentication and authorization. The library supports both symmetric and asymmetric cryptography, allowing for the use of shared secrets or public-private key pairs. It enables the creation of signed tokens and the verification of their authenticity and integrity to prevent data tampering. The toolkit covers a broad range of identity and access control capabilities, including the embedding of custom

    Decodes tokens and validates signatures to ensure payload integrity and authenticity.

    Goauthed25519go
    عرض على GitHub↗8,908
  • lastmile-ai/mcp-agentالصورة الرمزية لـ lastmile-ai

    lastmile-ai/mcp-agent

    8,037عرض على GitHub↗

    mcp-agent is a framework for building AI agents that integrate with Model Context Protocol servers to execute tools and access data. It functions as a multi-agent orchestrator and protocol-compliant server, enabling the creation of agents that can discover and invoke tools from connected external servers. The project distinguishes itself through a durable workflow engine that supports long-running tasks capable of pausing, resuming, and surviving restarts. It implements complex orchestration patterns, including iterative evaluator-optimizer loops, hierarchical workflow nesting, and specialist

    Handles the acquisition and storage of secure tokens to authenticate communication between agents and protocol servers.

    Pythonagentsaiai-agents
    عرض على GitHub↗8,037
  • mervinpraison/praisonaiالصورة الرمزية لـ MervinPraison

    MervinPraison/PraisonAI

    5,592عرض على GitHub↗

    PraisonAI is an autonomous AI agent platform that coordinates multiple LLM-powered agents for research, planning, and execution of complex workflows. It functions as a multi-agent orchestration framework, a workflow builder, and a Model Context Protocol server, while also providing retrieval-augmented generation through vector knowledge bases. Agents can interact via CLI, web, or standardized protocols with sandboxed code execution. The platform distinguishes itself with a rich set of agent communication protocols, including A2A, REST, WebSocket, voice and telephony integration, and MCP, allo

    Activates injection defense and audit logging across all agents with a single call.

    Pythonagentsaiai-agent-framework
    عرض على GitHub↗5,592
  • bc-security/empireالصورة الرمزية لـ BC-SECURITY

    BC-SECURITY/Empire

    5,045عرض على GitHub↗

    Empire is a post-exploitation command-and-control (C2) framework designed for red team operations. It deploys and manages agents written in PowerShell, Python, C#, Go, and C across Windows, Linux, and macOS, using encrypted communication channels over HTTP, HTTPS, and SMB. The framework executes over 400 built-in modules for reconnaissance, privilege escalation, credential theft, and lateral movement, and provides a modular engine for authoring custom attack modules. What sets Empire apart is its multi-language agent deployment system, which allows operators to choose implants that suit each

    Establishes encrypted communications between the control server and agents using a two-stage key exchange.

    PowerShellc2empirehacktoberfest
    عرض على GitHub↗5,045
  • microsoft/agent-governance-toolkitالصورة الرمزية لـ microsoft

    microsoft/agent-governance-toolkit

    4,522عرض على GitHub↗

    agent-governance-toolkit هو إطار عمل لفرض سياسات الأمان، وإدارة هويات الثقة الصفرية، وعزل تنفيذ وكلاء الذكاء الاصطناعي المستقلين. يوفر طبقة حوكمة مصممة للتحكم في سلوك الوكلاء من خلال استخدام محرك سياسات الأمان، وإدارة الهوية التشفيرية، وبيئة عزل تنفيذ وقت التشغيل. يتميز المشروع بنظام حلقة امتياز متعدد المستويات وشبكة هوية تشفيرية تؤمن الاتصال بين الكيانات المستقلة. ينفذ آلية تسجيل ثقة قائمة على التحلل لتتبع موثوقية الكيان ويستخدم سجلات تدقيق مشفرة ومضادة للتلاعب للحفاظ على تاريخ قابل للتحقق من التنفيذ. تغطي مجموعة الأدوات مجموعة واسعة من مجالات القدرات، بما في ذلك أمان المطالبات للدفاع ضد هجمات الحقن، ورسم خرائط الامتثال الآلي مقابل المعايير التنظيمية، وتنظيم سير العمل المستقل باستخدام أنماط الملحمة (Saga). كما يتميز بمراقبة الأسطول لتتبع الصحة وحدود الإنفاق، بالإضافة إلى عزل تنفيذ الأدوات لتقييد الوصول غير المصرح به للموارد. يتم توفير واجهة سطر أوامر لإرسال إشارات التحكم، والتحقق من سياسات الحوكمة، وإدارة تثبيت الامتدادات.

    Identifies and mitigates vulnerabilities in AI agents based on industry standards like the OWASP Agentic Top 10.

    Python
    عرض على GitHub↗4,522
  • panaversity/learn-agentic-aiالصورة الرمزية لـ panaversity

    panaversity/learn-agentic-ai

    3,908عرض على GitHub↗

    This project is an educational curriculum and architectural framework for building autonomous AI agents and multi-agent systems. It provides a structured learning path focused on the development of independent software components capable of planning, executing tasks, and utilizing external tools to achieve high-level goals. The framework emphasizes multi-agent system orchestration through distributed architectures where specialized agents collaborate using standardized communication protocols. It details specific design patterns such as dual-memory systems for maintaining short-term plans and

    Implements token and API key authentication to verify agent identity and authorize requests.

    Jupyter Notebooka2aagentic-aidapr
    عرض على GitHub↗3,908
  • velocidex/velociraptorالصورة الرمزية لـ Velocidex

    Velocidex/velociraptor

    3,769عرض على GitHub↗

    Velociraptor is a digital forensics and incident response platform, endpoint detection and response system, and visibility tool. It provides a query engine and remote forensic collector used to hunt for indicators of compromise and perform triage across a fleet of hosts. The system is distinguished by its specialized query language for interrogating host state and parsing binary files. It features a notebook environment that combines markdown documentation with executable query cells to standardize investigative workflows and enable collaborative reporting. The platform covers a wide range o

    Provides utilities to purge stale or inactive endpoint registrations from the server based on age thresholds.

    Godigital-forensicsendpoint-discoveryendpoint-protection
    عرض على GitHub↗3,769
  • datawhalechina/vibe-vibeالصورة الرمزية لـ datawhalechina

    datawhalechina/vibe-vibe

    3,126عرض على GitHub↗

    vibe-vibe is an LLM agent engineering framework and toolchain optimizer designed for orchestrating multi-agent systems. It serves as a comprehensive guide and methodology for transforming conceptual ideas into deployed applications through agentic software engineering. The project focuses on the orchestration of specialized AI agent roles with defined collaboration boundaries and iterative feedback loops. It provides frameworks for toolchain optimization, including the selection and evaluation of protocols that extend model capabilities and the design of standardized tool interfaces. The sys

    Determines appropriate isolation levels across tool and execution layers to secure autonomous agent systems.

    agentagentic-aiai
    عرض على GitHub↗3,126
  • moonshotai/kimi-codeالصورة الرمزية لـ MoonshotAI

    MoonshotAI/kimi-code

    2,473عرض على GitHub↗

    Kimi-code is a command-line interface and orchestration framework designed to integrate autonomous AI agents into software development workflows. It functions as a terminal-based assistant that manages multi-step coding tasks, including planning, file system modifications, shell command execution, and test running, all while maintaining conversational context within a local development environment. The project distinguishes itself through a focus on secure, autonomous agent orchestration and granular control over AI interactions. It enforces strict security by requiring explicit user approval

    Verifies agent identity using secure tokens for authorized operations on external platforms.

    TypeScript
    عرض على GitHub↗2,473
  • fetchai/innovation-lab-examplesالصورة الرمزية لـ fetchai

    fetchai/innovation-lab-examples

    1,028عرض على GitHub↗

    This project provides a comprehensive framework for building, deploying, and orchestrating autonomous agents within a decentralized network. It serves as a collection of patterns and examples for developing intelligent software entities capable of performing complex tasks, making decisions, and interacting with other agents to achieve shared goals. The framework distinguishes itself through its focus on multi-agent orchestration and decentralized communication. It enables the coordination of specialized agent teams that collaborate on workflows through structured messaging protocols, allowing

    Secures agentic architectures by restricting public exposure to a primary orchestrator while keeping workers local.

    Python
    عرض على GitHub↗1,028
  1. Home
  2. Security & Cryptography
  3. Agent Security

استكشف الوسوم الفرعية

  • Agent Registration UtilitiesCommand-line tools for automating the secure registration and provisioning of security agents. **Distinct from Agent Security:** Distinct from Agent Security: focuses on the registration and lifecycle management of agents rather than their internal security hardening.
  • Global Security PoliciesApplies injection defense and audit logging across all agents with a single configuration call. **Distinct from Agent Security:** Distinct from Agent Security: focuses on global application of security measures across all agents without per-agent configuration.
  • Isolation Layer DesignDesign of boundaries between tool and execution layers to prevent unauthorized agent actions. **Distinct from Agent Security:** Focuses on the structural isolation of layers rather than general security auditing of configurations.
  • Security Agent Inspectors1 وسم فرعيUtilities for retrieving status and configuration details of security agent instances. **Distinct from Agent Security:** Distinct from Agent Security: focuses on the inspection and health monitoring of agents rather than securing the agent architecture itself.
  • Security Agent RevocationsUtilities for removing and revoking access for registered security agents. **Distinct from Agent Security:** Distinct from Agent Security: focuses on the removal and lifecycle management of agents rather than securing the agent architecture.
  • Token AuthenticationMechanisms for verifying agent identity using secure tokens. **Distinct from Agent Security:** Focuses on token-based authentication for agent communication, distinct from general agent security auditing.
  • Two-Stage Key ExchangesEstablishment of encrypted communications between the control server and agents using a two-stage key exchange and encrypted routing. **Distinct from Agent Security:** Distinct from Agent Security: covers the two-stage key exchange for secure staging, not general agent security auditing.