Boucle-framework

Boucle Claude Code hooks that actually enforce your rules. 7 standalone hooks, plus enforce-hooks for CLAUDE.md policy, audit tooling, 1,900+ tests, and a searchable Claude Code gaps corpus with severity ratings and workarounds. Quick links: Check your setup · Install hooks · Known limitations ·...

cupcake

Cupcake Make AI agents follow the rules. Policy enforcement layer for AI agents; yielding better performance and security without consuming model context. Deterministic rule-following for your agents. Interactive Examples Better performance by moving rules out of context and into policy-as-code...

gatekeeper

Gatekeeper ¿En qué se diferencia Gatekeeper de OPA? En comparación con el uso de OPA con su sidecar kube-mgmt también conocido como Gatekeeper v1.0, Gatekeeper introduce la siguiente funcionalidad: Una biblioteca de políticas extensible y parametrizada CRDs nativos de Kubernetes para instanciar l...

kontext-cli

网站 | 文档 | 仪表盘 | Discord Kontext 是一个面向 AI 代理的授权平台。它帮助团队通过作用域凭据、策略执行、审批和审计跟踪来控制代理可以访问和操作的内容。Kontext 可以以本地优先的方式运行，适用于开发代理，并可扩展到受管或自托管部署，以满足安全敏感环境的需求。 快速开始 root@kitploit: brew install kontext-security/tap/kontext 将您的机器连接到工作区 使用自助设置将您机器上的代理活动流传输到团队的 Kontext 仪表盘。 在工作区的 Deployments 页面生成安装令牌，然后运行：...

agentsh

agentsh Nota para macOS: La aplicación nativa de enforcement en macOS mediante ESF Endpoint Security Framework + NE Network Extension está en Alpha. Funciona de extremo a extremo: los eventos de archivos, procesos y red fluyen a través de la extensión del sistema hasta el motor de políticas Go,...

prismor

Prismor 面向 Claude Code、Codex 及其他 AI 编码代理的运行时安全。 Prismor 可根据用户自身策略进行高度自定义。通过观察或强制模式，在本地自助式仪表盘中查看代理活动 网站 • 通过 Skill 快速上手 !Google Antigravityhttps://d205xtizsu1yjh.cloudfr...

opa

Open Policy Agent Open Policy Agent OPA es un motor de políticas de propósito general, de código abierto, que permite la aplicación unificada de políticas conscientes del contexto en toda la pila tecnológica. OPA se enorgullece de ser un proyecto graduado en el panorama de la Cloud Native Computi...

allstar

Allstar !IMPORTANT 由 OpenSSF 托管的 Allstar GitHub App 已退役。 Allstar 作为 OpenSSF Scorecard 子项目，其本身仍在持续维护中——您现在必须自行运行它，可以 作为 GitHub Action 或 作为服务守护进程 运行。 更多详情请参阅 ossf/allstar881。 如果您的组织一直依赖托管应用，请参阅 从托管应用迁移。 概述 什么是 Allstar？ Allstar 的新功能 whats-new.md 禁用不需要的问题 救命！Allstar 创建了我不想要的问题！ 入门指南 背景 组织级选项 安装选项 创建您...

agent-governance-toolkit

🌍 English | 日本語 | 简体中文 欢迎使用代理治理工具包 ! !IMPORTANT 公开预览版 — 此仓库中发布的所有软件包均为 经 Microsoft 签名的公开预览版 。它们达到 生产级别的质量，但在正式发布GA之前可能存在重大变更。如有任何反馈，请在GitHub上提交issue。 这个工具包是什么: 运行时治理基础设施 — 位于你的代理框架与代理执行操作之间的确定性 策略执行、零信任身份验证、执行沙箱，以及可靠性工程。 这个工具包不是什么: 这不是一个用于模型安全或提示词防护的工具。它不会过滤大语言模型 LLM的输入/输出，也不执行内容审核。它是在应用层对 代理的行为...

aegis-latent-core

🛡️ Aegis Latent Core AI Governance and Cryptographic Evidence Gateway Provider-independent request controls, bounded streaming PII redaction, and client-verifiable Merkle Mountain Range inclusion proofs for governed LLM traffic. Last verified: 2026-08-27 UTC Release baseline: che...

LID

██╗ ██╗██████╗ ██║ ██║██╔══██╗ ██║ ██║██║ ██║ ██║ ██║██║ ██║ ███████╗██║██████╔╝ ╚══════╝╚═╝╚═════╝ Linux Integrity Drift — "Linux está muriendo" — Descubrimiento sistemático de rutas de código del kernel que omiten las garantías de seguridad de LSM La puerta nunca fue derribada. Simplemente la...

aegisagent

AegisAgent — AI Agent Runtime Security Gateway Puerta de enlace de seguridad en tiempo de ejecución para AI Agent: vincula criptográficamente cada llamada de herramienta a la finalización real del modelo, con ejecución de políticas de confianza cero, aislamiento en sandbox y aprobación humana...

CVE-2026-82474

Sudo through 1.9.17p2 fails to apply intercept policy checks to the execveat system call in ptrace-based intercept mode. A local user with permission to run specific commands can execute denied programs by calling execveat directly or via fexecve , thereby bypassing both policy enforcement and au...

CVE-2026-82474 Sudo through 1.9.17p2 Intercept Policy Bypass via execveat

Sudo through 1.9.17p2 fails to apply intercept policy checks to the execveat system call in ptrace-based intercept mode. Users permitted to run specific commands can execute denied programs by calling execveat directly or through fexecve, bypassing policy enforcement and logging...

conductai

Conduct Governança em tempo de execução para agentes de IA — uma política aplicada em cada chamada de LLM, em cada ferramenta de shell, em cada sessão de IA dos colegas de equipe. Duas superfícies de produto, um repositório, uma política: Conduct Guard — o motor de políticas. Decide block / warn ...

CVE-2026-19286 Langflow is affected by multiple remote code execution vulnerabilities due to insufficient code-execution policy enforcement

IBM Langflow OSS 1.0.0 through 1.11.1 could allow a remote attacker to execute arbitrary code due to improper enforcement of security restrictions on the A2A public endpoint...

CVE-2026-19286

IBM Langflow OSS versions 1.0.0 through 1.11.1 are vulnerable to arbitrary code execution via the A2A (Agent-to-Agent) public endpoint . The root cause is that the A2A public execution path does not invoke the validate_public_flow_no_code_execution guard used by other unauthenticated public surfa...

CVE-2026-18729

IBM Langflow OSS versions 1.0.0 through 1.11.1 are affected by a code injection flaw (CWE-94). The PythonFunction component lacks the runtime code-execution gate applied to other components, allowing a remote authenticated attacker to bypass the allow_custom_components=false hardening policy via ...

CVE-2026-18729 Langflow is affected by multiple remote code execution vulnerabilities due to insufficient code-execution policy enforcement

IBM Langflow OSS 1.0.0 through 1.11.1 could allow a remote authenticated attacker to execute arbitrary code due to improper control of generation of code...

Astra Linux – Vulnerability in Chromium

Insufficient policy enforcement in extensions in Google Chrome prior to 88.0.4324.96 allowed a remote attacker to bypass site isolation through a malicious Chrome Extension...