The scary 2026 stat isn't the 340% surge in prompt injection or the 88% of orgs reporting agent...
The scary 2026 stat isn't the 340% surge in prompt injection or the 88% of orgs
reporting agent incidents (OWASP-linked, Beam AI). It's this: the leak
is often not in the answer your logs capture — it's in the model's reasoning, which
most people never scan. A 2026 benchmark found data escaping through reasoning/logs
while the visible output stayed clean in most scenarios (AgentLeak).
I built a scanner that fires prompt-injection probes at a self-hosted AI agent and checks whether it...
I build a small open-source tool (rojaprove) that checks whether an AI app leaks its hidden...
(Early, single-configuration findings — read the caveats before you quote a number.) If you ship a...
How I built the first open-source tool to catch the attacks that no guardrail sees coming We talk...
Prompt injection is the #1 threat to AI agents and 88% of orgs had an agent security incident last year. The bottleneck in…
The pull request looked clean. The tests were green, the diff was small, and the commit message was...
