Michael Cutler

Fractional Futurist

~/blog/tag $ grep -r "#ai-security"

#ai-security

2 posts

May 30, 2026|5 min read

The 30 Principles for Agentic Engineering — Part 4: Governance and Safety

Principles 21–25. The governance and safety layer: strictKnownMarketplaces, no goal-conflict prompts, quarterly AppSec, four telemetry signals, monthly incident discipline.

[AI & Data][Security]+1

May 15, 2026|6 min read

Snyk's ToxicSkills Audit: 13.4% of Public Skills Are Vulnerable

Snyk audited 3,984 public Claude Code skills. 13.4% had critical vulnerabilities. 76 were confirmed malicious. ClawHavoc is the separate, scarier story. Here's the supply-chain hygiene most teams aren't doing.

[AI & Data][Security]

Preparing your content