mitos

Mitos Computadoras aisladas y bifurcables para tus agentes de IA. Bifurcación de sandbox microVM en milisegundos en Kubernetes: bifurca una VM en ejecución en intentos paralelos y restáurala desde la memoria en decenas de milisegundos. Inicio rápido . Documentación . Características . Comparación...

nono

由打造 Sigstore 的团队构建 安全软件证明的标准，被 PyPI、npm、brew 和 Maven Central 采用 !NOTE 在迈向 1.0 版本发布的过程中，API 正在趋于稳定。必要时仍可能发生 API 变更，但会保持在最低限度。 !IMPORTANT 组织迁移： 官方 nono registry 命名空间已从 always-further 迁移到 nolabs-ai。请更新脚本、配置文件和 CI 中的所有引用。 如果您已经在旧命名空间下安装了包，请先移除，再从新的命名空间拉取： root@kitploit: nono remove...

AgentGuard

AgentGuard 面向自主AI代理的零信任沙箱。 AgentGuard 可以用安全护盾包裹任何 AI 代理（LangChain、CrewAI、AutoGen、自定义脚本）。只需改变一个命令： root@kitploit: 之前（危险——代理具有完全系统权限） python myagent.py 之后（沙箱化） agentguard run -- python myagent.py AgentGuard 拦截代理发出的每条 shell 命令、文件修改和网络请求。安全操作自动放行，危险操作自动拦截，其余操作则提示人类审批。 工作原理 AgentGuard 有四层防御层协同工作：...

coderunner

CodeRunner: Un entorno aislado local para tus agentes de IA CodeRunner te ayuda a aislar tus agentes de IA y sus acciones dentro de un entorno aislado. Caso de uso clave: Puedes ejecutar múltiples Claude Code o agentes de IA en nuestro entorno aislado sin temor a pérdida de datos o exfiltración...

yolo-cage

yolo-cage: 不造成伤害的自主编码代理 你是一位负责任的工程师。你绝不会让AI在你的最敏感系统和代码库中随意横行。 这就是为什么你绝不会 仅仅关闭像Claude Code这样的工具的安全保护。它会对每个危险操作询问许可！安全！ 于是你等待。你回答。决策疲劳开始产生。然后那一刻发生了。 权限提示忽略了威胁模型中最薄弱的部分：疲惫的用户。如果我们能赋予代理权力同时限制其影响范围，从而将你的决策延迟到PR审查时，那会怎样？ 那太好了！这就是 yolo-cage。 尝试它 root@kitploit: curl -fsSL...

CVE-2026-42302 FastGPT: Unauthenticated Remote Code Execution (RCE) via code-server Misconfiguration in agent-sandbox

FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the agent-sandbox component of FastGPT is vulnerable to unauthenticated Remote Code Execution RCE. The startup script entrypoint.sh initializes code-server with the --auth none flag and binds the service to...

CVE-2026-42302

FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the agent-sandbox component of FastGPT is vulnerable to unauthenticated Remote Code Execution RCE. The startup script entrypoint.sh initializes code-server with the --auth none flag and binds the service to...

CVE-2026-42302 FastGPT: Unauthenticated Remote Code Execution (RCE) via code-server Misconfiguration in agent-sandbox

FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the agent-sandbox component of FastGPT is vulnerable to unauthenticated Remote Code Execution RCE. The startup script entrypoint.sh initializes code-server with the --auth none flag and binds the service to...

CVE-2026-42302 FastGPT: Unauthenticated Remote Code Execution (RCE) via code-server Misconfiguration in agent-sandbox

FastGPT is an AI Agent building platform. From version 4.14.10 to before version 4.14.13, the agent-sandbox component of FastGPT is vulnerable to unauthenticated Remote Code Execution RCE. The startup script entrypoint.sh initializes code-server with the --auth none flag and binds the service to...

CVE-2026-42302

FastGPT: agent-sandbox vulnerable in 4.14.10–4.14.12 due to entrypoint.sh launching code-server with --auth none and binding to 0.0.0.0:8080, enabling unauthenticated remote code execution and full sandbox access. The issue is mitigated in version 4.14.13. Practical impact is unauthenticated netw...

PT-2026-39205

Name of the Vulnerable Software and Affected Versions FastGPT versions 4.14.10 through 4.14.12 Description The agent-sandbox component allows unauthenticated Remote Code Execution RCE, which is the ability to execute arbitrary commands on a remote machine. The startup script entrypoint.sh...

PYSEC-2026-106

OpenHands is software for AI-driven development. Starting in version 1.5.0, a Command Injection vulnerability exists in the getgitdiff method at openhands/runtime/utils/githandler.py:134. The path parameter from the /api/conversations/conversationid/git/diff API endpoint is passed unsanitized to ...

CVE-2026-33718 OpenHands is Vulnerable to Command Injection through its Git Diff Handler

OpenHands is software for AI-driven development. Starting in version 1.5.0, a Command Injection vulnerability exists in the getgitdiff method at openhands/runtime/utils/githandler.py:134. The path parameter from the /api/conversations/conversationid/git/diff API endpoint is passed unsanitized to ...

CVE-2026-33718

OpenHands is software for AI-driven development. Starting in version 1.5.0, a Command Injection vulnerability exists in the getgitdiff method at openhands/runtime/utils/githandler.py:134. The path parameter from the /api/conversations/conversationid/git/diff API endpoint is passed unsanitized to ...

CVE-2026-33718 OpenHands is Vulnerable to Command Injection through its Git Diff Handler

OpenHands is software for AI-driven development. Starting in version 1.5.0, a Command Injection vulnerability exists in the getgitdiff method at openhands/runtime/utils/githandler.py:134. The path parameter from the /api/conversations/conversationid/git/diff API endpoint is passed unsanitized to ...

PT-2026-28181

Name of the Vulnerable Software and Affected Versions OpenHands versions prior to 1.5.0 Description OpenHands is software for AI-driven development. A Command Injection vulnerability exists in the get git diff method at openhands/runtime/utils/git handler.py:134. The path parameter from the...

LLM Agents Should Employ Security Principles

Large Language Model LLM agents show considerable promise for automating complex tasks using contextual reasoning; however, interactions involving multiple agents and the system's susceptibility to prompt injection and other forms of context manipulation introduce new vulnerabilities related to...

PT-2026-2468

Name of the Vulnerable Software and Affected Versions Node.js affected versions not specified Description A flaw exists in the Node.js software platform due to incorrect path name restriction for restricted access directories. Successful exploitation of this issue could allow an attacker to...