Writing
Field notes from the studio.
Offensive security, AI red-teaming, and the compliance deadlines that now come with shipping a model. Mechanisms rather than command dumps, and every attack paired with how you would catch it. The full taxonomy lives in the field manuals; this is where we work in the open.
A note from the founder
Why we are giving the writing and the audit tool away for free, and what I am actually trying to build here — a founder's note from Komal Mishra.
The lethal trifecta — why your agent is one email away from exfiltration
An LLM agent becomes dangerous the moment it holds three things at once: private data, exposure to untrusted content, and a way to talk to the outside world. Here is why that combination leaks, and what to actually remove.