Latest / Decrypted: The UK Cyber Briefing

A forged GitHub comment, and the coding agents that couldn't tell the difference
Morning edition, Monday 20 July 2026
New research shows AI coding and browsing agents can be fooled by forged metadata rather than obvious prompt injection, exactly the risk NCSC guidance warned about in May. Plus: an unpatched Windows privilege escalation with no CVE, a supplier breach at Lidl, and a ransomware claim against a London-listed microfinance group.
Sources
Agent Data Injection Attacks are Realistic Threats to AI Agents (arXiv:2607.05120)
New Agent Data Injection Attack Can Make AI Agents Misclick or Run Attacker Commands - The Hacker News
Thinking carefully before adopting…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.