OpenAI staffer pitches agent inbox, memory kill-switch and local computer use for ChatGPT
willdepue · x · 2026-10-03
OpenAI employee willdepue posted a public wishlist for ChatGPT and Codex:
- Agent inbox: a Kanban-like multi-inbox view aggregating tasks and questions from agents, instead of scrolling through project chats — he argues it would let him manage 10x more agents without losing track of old threads.
- ChatGPT Dot transparency: surprised that Dot can't see or manage his Codex projects; wants it to handle all his tasks and projects, plus local computer use.
- /disable-memory command for ChatGPT and Codex to skip memories/instructions on single turns without going incognito.
- He is also hacking together integrations bridging agents to iMessage, Signal, WhatsApp, Telegram and X DMs.
Related event: OpenAI Employee Pitches Agent Inbox, Memory Toggles and Model Presets(2 posts)→
More from coding & agent
- FT: US tech outages blamed on rush to fix bugs found by AI — nordicinst · 2026-11-19
- Claude Code overtakes Cursor and Copilot in job-post mentions ahead of Anthropic IPO — antgoldbloom · 2026-11-19
- Dev submits PR adding periodic checkpoints to GKE sandboxes so agents don't lose progress — yoimnotkesku · 2026-11-19
- AWS adds private CA support to Bedrock AgentCore Identity for connecting agents to self-hosted IdPs — AWS ML Blog · 2026-11-19
- Orchestrator's Cloud Agents graduate from beta to general availability — julianweisser · 2026-11-19
- Zero users, 33 pages shipped: dev crowdsources audit for AI-agent trust schema — TheHamzaShahbaz · 2026-11-19