Blog.

AI agent security, CVE breakdowns, and infrastructure for autonomous systems.

February 23, 2026 // SECURITY SUPPLY CHAIN

36% of ClawHub skills have security flaws. I built a scanner.

Snyk scanned 3,984 agent skills and found 76 confirmed malicious payloads: credential theft, backdoor installation, remote code execution. Agent Passport v2.3.0 adds pre-install scanning and runtime injection detection.
READ POST →
February 22, 2026 // SECURITY CVE

Agent Passport blocks every recent OpenClaw CVE. Here is how.

OpenClaw had nine vulnerabilities disclosed in under a month: SSRF in the gateway, missing webhook authentication, path traversal in the browser upload handler, and a one-click RCE. Here is what each one does, and what blocks it.
READ POST →