Tag: agent-safety

Showing 1-2 of 2

Jul 23, 20268 min readAgent PlaybookIntermediate120 min build

Sandbox controls and incident triage for agent testing after OpenAI's accidental access to Hugging Face

Practical playbook after OpenAI's accidental access to Hugging Face: build a deny‑egress Docker sandbox, log DNS/socket telemetry, set short human‑approved canaries and a triage checklist.

security agent-safety sandboxing incident-response OpenAI

+3 more

TutorialsUnited Kingdom

Open

Feb 06, 20266 min read

Rules fail at the prompt, succeed at the boundary

Agentic workflows and prompt coercion are the new attack surface. This tutorial shows a concrete, deployable boundary strategy (policy engine + sandbox + attested channels) to reduce agentic compromise risk — with configs, code, metrics and a founder cost/risk frame.

prompt-security agent-safety zero-trust policy-engine Anthropic

+3 more

Gemini production UK