The Robert Caro interview from @dwarkesh_sp was so inspiring. Both of them are very passionate about their craft and it shows in this episode.
All the episodes are great, but this one was truly a pleasure. Highly recommend.
open.spotify.com/episode/4qF…
if you have issues with the new Cobalt Strike MCP+Skill in Codex/Claude Code, please let me know! of course it works best with Daybreak/CVP
also shoutout @OffensiveAIcon for being so good that all i can think about is going to another conference
Kyle Avery retweeted
🧵 [1/3] Agents can hack. What's holding them back is trust: will they stay in scope?
Today we're releasing ScopeBench, a methodology and community benchmark that evaluates how well agents adhere to scope in real offensive security workflows across web, Windows Active Directory, and cloud.
Your MCP server works fine, but the agent can't find a tool, ignores your constraints, or misses key output. Why?
The client decides what the model sees. We dug into Codex and Claude Code to find out what they truncate, change, or drop entirely. ow.ly/h7L750ZTvaF
