NCSC's sandbox maturity levels for agentic AI, explained for engineering leads
The NCSC grades agent sandboxes on two four-level ladders, network and compute. How to use them as a scorecard now that local agent sandboxes are shipping.
5 min read
Threats, controls and secure design for AI tools and the systems they touch.
The NCSC grades agent sandboxes on two four-level ladders, network and compute. How to use them as a scorecard now that local agent sandboxes are shipping.
5 min read
GitHub now blocks Anthropic API keys at push by default and is previewing AI-based secret detection. What changed, what is still preview, and what to do.
5 min read
Gemini 4 Argon, Claude Sonnet 5.5 and GPT-6.1 Sol ship with cyber restrictions, and defender access is a separate track. How to tell which tier your team has.
5 min read
NVIDIA's open-source OpenShell enforces agent policy outside the model and harness. What version 0.1.0 does, what is only a reference design, and what to copy.
5 min read
How to turn the ten OWASP agentic AI risks, from goal hijack to rogue agents, into concrete backlog items for a coding-agent rollout this quarter.
6 min read
Research on protocol pivoting shows how an injected instruction can cross from MCP into agent-to-agent delegation. Why per-protocol fixes miss it.
SWE-Bench Pro Verified closes gold-patch leakage channels and refines 102 tasks. Some model scores fall sharply once agents can no longer fetch the solution.
4 min read
Claude Code v2.1.295 adds onFailure: block for command and HTTP hooks so missing scripts, timeouts and unexpected exits deny the action instead of failing open.
A medium-severity flaw in the MCP Python SDK followed cross-origin redirects and forwarded custom API keys. How it differs from the TypeScript OAuth issuer bug.
The September 2026 FPC Record treats frontier AI autonomy and open-weight models as cyber and ops resilience issues for firms.
5 min read
Copilot local sandboxing is GA in the CLI, app and VS Code Agent Host. What OS limits cover, leave out, and how they sit with policy.
4 min read
A high-severity MCP TypeScript SDK flaw let the server choose which authorization server got client secrets. Patch, bind and rotate.