TL;DR We added architectural context to AI coding agents via MCP and tested on SWE-bench...
How a $0.02/Call Model Scored 78.2% on SWE-bench Verified — Beating Every Model on the Leaderboard
Related
Agent Protocols Fail at the Seams | Focused Labs
MCP, A2A, and ACP simplify agent integration, but security ownership sits in the runtime where content, authority, and state cross protocol seams.
Code Review, Part 2: The Reviewer That Learned To Lie Better
Several posts ago, I wrote about setting up a multi-agent adversarial code review process as part of...
I asked an AI agent to delete a folder my tool was guarding. Here's everything that went wrong.
I build Termaxa, a small Rust CLI that gates the shell commands AI coding agents run: it previews...