Building a Warehouse Shift Handoff with Six AI Agents in Switch
How we used a shared Telegram room to organize incident evidence, map operational impact, and keep people in control. At shift change, an operations team needs clear answers: What is confirmed? What is still uncertain? Which work is blocked? Who has accepted the next action? We built a warehouse shift-handoff workflow with Switch, Telegram, and six role-based AI agents. The goal is to make…
A warehouse shift handoff was created using a shared Telegram room, Switch, and six role-based AI agents. The goal was to make incident context easier to review and carry over from one shift to the next. The warehouse incident in the demo was simulated and not connected to a live system. Switch connected Telegram to the agents and kept their findings visible alongside human conversation.
Six agents had specific responsibilities: Statekeeper maintained the current state, Conflict Hunter identified contradictions, Dependency Mapper traced downstream impact, Risk Forecaster described operational risks, Ownership Agent recommended next-shift owners, and Auditor ensured handoff completeness. The shared room instructions prevented agents from inventing details, preserving sources, and marking unresolved facts as unknown.
A simulated incident involved conveyor C3 stopping at 09:10 and maintenance ticket M-204 being open. O-81 and O-82 were confirmed blocked due to C3's stop, and C4 had no spare capacity before 15:00. After adding follow-up evidence, the agents updated their findings. One risk message was corrected by a human to reflect evidence accurately.
The handoff remained incomplete as repair clearance and maintenance ownership were unresolved. Clear agent roles, evidence rules, pending ownership, and visible unresolved questions were learned from this prototype.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
