User's open letter to OpenAI: Dot agent degraded after a week, switching to Claude Duet
Danny7Guns · reddit · 2026-11-17
A Reddit user published an open letter criticizing OpenAI's Dot agent: excellent for its first week, then progressively worse — forgetting SOPs, violating workflow rules even after reminders, and suffering shrinking memory. They patched around it with agents.md, config.toml, plugins, third-party memory tools, and scheduled-task workarounds.
They praise Anthropic's Claude Duet for isolated multi-session project contexts, which Dot lacks when multitasking. Their verdict: OpenAI shipped Dot before it was ready — "no point releasing a new engine when the powertrain isn't working" — and they'd rather see Dot fixed than new 6.2 models.
More from coding & agent
- Microsoft open-sources TokenEconomizer to cut coding agents' token waste — udmrzn · 2026-11-17
- Dev Burned 9 Billion Tokens Last Month, So He Turned Claude Code Usage Into a Battle Game — Fun_Craft_3089 · 2026-11-17
- Dev uses Claude Code to reverse-engineer a cheap AI badge into an open-source Claude voice assistant — ConsequenceThen217 · 2026-11-17
- Codex lacks visual indicator for multi-day agent runs, user resorts to renaming tasks — chrisalbon · 2026-11-17
- Papertrails: An Endless Papercraft World Built in Browser with AI — CatAstro_Piyush · 2026-11-17
- J.Doe learns your preferences as rules and hands work to Claude Code and Codex agents — eyishazyer · 2026-11-17