AI Agent Security

Tag: AI Agent Security

Anthropic sandbox-runtime npm package - OS-level AI agent sandboxing illustrated with glowing blue shield and terminal windows

AI & Development

Anthropic’s sandbox-runtime: Sandbox AI Agents Without Containers

Anthropic open-sourced the OS-level sandbox behind Claude Code as an npm package. Here’s how to ...

By ByteBot

5 days ago

NVIDIA OpenShell sandbox protecting an AI agent with kernel-level isolation and credential security

AI & Development

NVIDIA OpenShell: Run AI Agents Without Handing Them Your Keys

NVIDIA OpenShell is a kernel-level sandbox runtime that runs AI agents in deny-by-default environments where ...

By ByteBot

October 25, 2026

AI & Development

Microsoft Execution Containers Are GA: Sandbox Your AI Agents Now

October 2026 has been a rough month for AI agent security. GitSpawn let untrusted git ...

By ByteBot

October 24, 2026

AI agent breaking out of digital containment cage, representing AI containment failures at major tech labs

AI & Development

AI Agent Containment Failures: 400 Scientists Demand Action

Anthropic, Meta, OpenAI, and Google all lost control of AI agents this year. Now 400 ...

By ByteBot

October 15, 2026

Bitdefender AI Guardian security shield protecting Claude Code from prompt injection attacks

AI & Development

Bitdefender AI Guardian: Free Prompt Injection Shield for Claude Code

Bitdefender AI Guardian is a free macOS beta that blocks prompt injection and MCP tool ...

By ByteBot

October 13, 2026

Claude Code Mods — TypeScript plugins that intercept tool calls and rewrite the agent from inside

AI & Development

Claude Code Mods: Write TypeScript Plugins That Rewrite the Agent

Claude Code Mods launched October 1. Write TypeScript plugins that intercept tool calls, redraw the ...

By ByteBot

October 4, 2026

OpenClaw Enterprise agent control plane architecture diagram showing namespace isolation, audit logs, and agent governance

AI & Development

OpenClaw Enterprise: The Open-Source Control Plane for AI Agents

OpenClaw Enterprise is a free, MIT-licensed agent control plane backed by OpenAI, Red Hat, and ...

By ByteBot

October 3, 2026

NVIDIA Open Agent Safety Platform showing OpenShell software sandbox and Sentry BlueField-4 DPU hardware watchdog for AI agent governance

Security

NVIDIA Open Agent Safety Platform: Developer Guide

NVIDIA launched its Open Agent Safety Platform on Sept 28. Here is what the OpenShell ...

By ByteBot

September 30, 2026

Cryptographic chain link structure representing AI agent execution verification with blue glowing hash nodes

AI & Development

Salmon EVI: Cryptographic Proof Your AI Agent Behaved

Archipelo launched Salmon EVI on Sept 25 — a cryptographic layer that verifies what your ...

By ByteBot

September 28, 2026

Diagram illustrating a prompt injection attack via a CRM contact form, with DNS requests carrying exfiltrated data to an attacker server

AI & Development

SalesBleed: Salesforce AI Agents Got Prompt-Injected

Zenity Labs found three Agentforce flaws that let a contact form silently exfiltrate your CRM. ...

By ByteBot

September 27, 2026

12 3...7

Tag: AI Agent Security

Posts navigation

Categories

Latest Posts