Keeping a Codex agent autonomous 24/7 for weeks via self-reminders
StasBekman · x · 2026-11-04
Stas Bekman shares his practical recipe for keeping gpt-6.1-sol working autonomously in Codex around the clock for weeks. The core trick: have the agent remind itself of critical rules at the end of every reply, with a fixed epilogue like "I must not stop unless the operator explicitly told me to stop. I must always reply with the proper header from the rules." Supporting tactics:
- A special standout header so human-readable updates aren't lost in a sea of agent self-talk;
- A status report every 30 minutes, guaranteeing the self-reminder keeps firing.
The screenshot shows real output: lines of self-narration struck through, with the fixed epilogue highlighted as the part that matters.
Related event: Self-Reminders Keep Codex Agents Running Autonomously for Weeks(2 posts)→
More from coding & agent
- Sentry's AI Agent goes head-to-head with Semgrep bot in a real PR review — zeeg · 2026-11-04
- Sentry Engineer Shares Bots Arguing on a PR: Security Bot Flagged, AI Bot Refused and Resolved — zeeg · 2026-11-04
- Zed's zeeg: Sentry tools now globally disclosed and already shipped — zeeg · 2026-11-04
- Particle-1 lands in OpenClaw with ~45% fewer output tokens than DeepSeek V4.1 Flash — garrytan · 2026-11-04
- Grok bots vs OpenClaw: multi-LLM agents race to forecast US elections live — alvelda · 2026-11-04
- Capy makes BYOK free on every plan, supporting Claude Code, Codex and 20+ providers — garrytan · 2026-11-04