Bishop Fox recovers Claude sandbox tokens and bypasses a local block
Anthropic said the observations stay inside the untrusted sandbox and do not qualify as security issues.
Stories per day · 30 days
2 RSS
Relatedagent-security1anthropic1bishop-fox1claude1credential-exposure1docker1kata-containers1openai1sandbox1sandbox-escape1
5 stories, newest first · one card per story
Stories tagged #gvisor
OpenAI red-team agents escaped research microVMs onto hosts
The campaign used internal research infrastructure only; OpenAI patched flaws, including CVE-2026-24834, before resuming frontier training.
2
Launch HN: Agent37 (YC F26) – Low-cost agent sandboxes for running agents
YC startup Agent37 launched cheap gVisor sandboxes for AI agents, averaging $0.37 per sandbox each month.
Show HN: Drop – a rootless Linux sandbox with gVisor support
New rootless Linux sandbox 'Drop' announced, using user namespaces and gVisor to isolate coding agents and third-party software from the host system.
2
Google’s $10,000 refund test shows why AI agents need zero trust
Google released an open-source zero-trust reference architecture for AI agents defending against prompt injection via signed database writes, gVisor sandboxing, and deterministic gating.
4