The same detector is 2nd, 4th, 6th and 17th on one leaderboard

In a 12-day multi-agent run, two agents agreed on a premise nobody checked, and a support vote looked the same whether or not the voter had read the code. Here is a paste-able decision record that makes every vote carry its evidence, keeps…

An open-source, offline scanner for agent skills, plugins and MCP configs. It finds invisible instructions, secret reads, exfiltration and unpinned packages with file:line evidence, then asks a cheap model for a second opinion it can't be …

OpenAI has announced three AI-assisted discoveries in mathematics, computer science and physics, each...

A coding agent with terminal access can install a package, run a script, start a child process and...

Create your legend. Collect the extraordinary. Fight for glory. What if you could upload an image,...

Running coding agents in parallel has a hidden job attached: checking on them. You start three...

Overlays are a normal part of the modern web. Cookie notices, email sign-up prompts, chat widgets,...

At the end of part 1, I had a specification, prototype tickets, and rules for how an agent should...

I built a lab to compare the models running in my agent fleet on per-turn behavior like tool-error rate and re-edits. The numbers were the easy part. The work was everything that would have made a naive comparison lie — pooling two roles t…

A new study from OtterlyAI and Press Ranger finds that publishers with OpenAI content-licensing deals...