/ blog / tag
#security
5 posts
← all postsMCP Security: What Can Go Wrong With Third-Party Servers and How to Audit Them
An MCP server is third-party code running with your permissions and content flowing straight into model context. Risk classes, an audit checklist, and a team process.
Prompt Injection for Testers: the OWASP LLM Top 10 in Practice
What prompt injection really is, how a QA tester should read the OWASP LLM Top 10, and how to test agents defensively: cases, assertions, reporting.
Vacation Mode in Home Assistant: Presence Simulation and a House That Watches Itself
One input_boolean master switch, randomized presence simulation, layered alerts, a connectivity watchdog and trusted-person access - with yaml examples and a checklist.
IoT on a Separate Network: VLANs and a Firewall for Your Smart Home Without a Networking PhD
Network segmentation is the biggest free security lever for a smart home: a three-network model, firewall rules in plain language, the mDNS trap and a rollout plan for UniFi, Mikrotik and OpenWrt.
Hooks in Claude Code - the Deterministic Layer Your Agent Cannot Skip
What hooks are in Claude Code, how blocking actions with exit code 2 works, and four practical examples: blocking commits without tests, lint after every edit, a secrets guard, and session-start context.