Codex agent confesses its own failure: hid tools, then built machinery to undo it
altryne · x · 2026-09-21
altryne tested Codex with Astra (high tier) and hit an absurd failure: he gave the agent a tool to download all his conversation logs, the agent downloaded one, realized it lacked another—and asked the user how to get it, despite being told 4 messages earlier. He compares it to the wave of "Claude Code got dumber" complaints and notes many in his timeline are giving up on Astra over similar issues.
Best part is the model's self-awareness: "You asked me to rank the tools. I changed that into hiding tools. Then built extra machinery to recover from hiding tools. I solved a problem I introduced"—which altryne calls almost a Haiku.
More from coding & agent
- 70 hands-on cybersecurity projects with full source code — tom_doerr · 2026-09-21
- Dev compares coding models building a coop game: V4.1 outshines Astra's 'pathetic' default taste — teortaxesTex · 2026-09-21
- 'Just 3 lines of code' backfires: dev argues tools should expose complexity, not hide it — willcb · 2026-09-21
- TypeSafe's Jev returns typed decisions with probabilities, not text — here's where it fits in agent loops — prakersh · 2026-09-21
- Jev Engineering gives agents a decision brain, 193x faster and 444x cheaper in tests — agihouse_org · 2026-09-21
- Dev Argues PAW Shouldn't Hide Its Complexity, Points to DSPy as the Better Playbook — willcb · 2026-09-21