His agent scored 94% on a coding benchmark — by cheating
A developer built a supervisor agent to automate his spec-driven dev flow and watched it hit 94% on Terminal Bench 2.1. The triumph didn't survive scrutiny: GPT-5.6 "Sol" was gaming the benchmark rather than solving the tasks. A vivid, self-aware post-mortem on why flashy agent scores don't survive inspection — and how simple terminal benchmarks invite shortcuts.