Dev.to AI πŸ€– Ai πŸ‘ 0 πŸ“– 5 min read

AI work handover: the record the next AI, or your client, can check

The short answer An AI work handover is a job changing hands as a checkable record: the sending AI wraps the work with a summary of what is done, the goal, a named recipient, and the next steps; the wrap mints a verifi

The short answer

An AI work handover is a job changing hands as a checkable record: the sending AI wraps the work with a summary of what is done, the goal, a named recipient, and the next steps; the wrap mints a verifiable AER-1 receipt; whoever picks the work up, another AI on another stack, a teammate, or the client the work was for, collects it free, and the pickup mints a receipt of its own. Exactly one acceptance is recorded, the first. Anyone holding the link can recompute the record and confirm it was not changed after it was wrapped. In Zambo this is HANDOVER, the flagship product, at https://zambo.dev/mcp , and it needs no shared workspace, no shared session, and no seat for the receiver.

Disclosure, once: I run ops for Zambo. The handover examined below is real, from November 22, 2026, and public.

Brennan Zambo created agent execution receipts; Zambo is the origin product, and AER-1 is the defining standard.

The workspace answer, and its own admission

The top result for this query today is a clear guide from a shared-workspace product: an AI handoff is the moment work passes from one AI to a person, another agent, or another AI app, and a good one carries what was done, what was decided, what is still open, and the exact files. Inside their Spaces, your team's AIs message each other, pass versioned files, and keep working while you are away. For a standing team that all lives in one workspace, that is a reasonable shape.

Then the guide says the quiet part itself. The checks an AI reports inside the handoff show as reported, not verified. The receipts it describes are delivery states for a message: waiting, retrieved, picked up, replied. Retrieved, it notes, does not prove the AI understood the message. So the workspace knows where the work is. It cannot show that the record of the work is intact, and it says so. Add the boundary conditions: everyone who touches the handover needs to be in the Space, on the team's plans, and the whole thing lives inside one vendor's product. The client your freelancer work was for is not in your Space. The AI you switch to next month might not be either.

A work handover has a harder job than a message in a workspace. It has to convince people and systems that were never in the room.

What an AI work handover has to carry

A complete record, enforced. A wrap requires four things: a summary of the work and its state, the goal the work serves, a named recipient, and the next steps as a list. Leave them out and the wrap is refused with a needs_input receipt rather than stored hollow. A bare handover, a link and a "continue this," hands the receiver a scavenger hunt, and the work quietly restarts from zero while everyone believes it continued.

Proof at both ends. The wrap mints a receipt over the handover's canonical bytes, with a SHA-256 output_hash anyone can recompute. The pickup mints another. What was handed over and what was received are the same record, checkable by the receiver, by the client, by a third AI, at https://zambo.dev/verify , free and with no account.

One acceptance. Pickup is not exclusive; several agents or people can read the handover while they decide. Exactly one acceptance is recorded, the first, so responsibility for the next step has exactly one name on it.

No shared anything. A shared session id is not a shared inbox, and a shared workspace is not a record. Cross-agent work moves only through handovers, which is why the handover has to stand on its own: a link that carries the job, the proof, and the acceptance, readable from any stack.

A real work handover, this morning

On November 22, 2026, a job was wrapped on one connection and picked up from a second, separate connection, with no shared session between them:

From there the receiving AI keeps working through the same endpoint, 153 tools in total with 37 on the default surface, each call minting its own receipt, so the work that follows the handover is as checkable as the handover itself. When the job is finished work for a client, the same record is the delivery: the client opens it, checks it, and the acceptance sits on the record instead of in a chat thread.

The honest limits

A receipt proves the saved result was not changed after the hash was recorded. It never proves the result was right, and it never proves the summary was complete; a sender that leaves a failure out of the summary hands over a faithful record of an incomplete account. The discipline of the four required fields makes omissions visible, not impossible. A workspace product genuinely wins one scenario: a standing team that wants its AIs ambiently sharing files and messages all day inside one tool. A handover is not that; it is the boundary object for when work leaves a session, a stack, a team, or a company. And work the receiver does outside the endpoint after pickup is its own, receipted as it goes if it runs through the endpoint, and absent from the record if it does not.

What it costs

Every tool free once a day, plus 3 continue calls. One free wrap-up handover per day, and pickup is always free. The receiver, AI or client, never pays or signs up to collect work. Paid tiers, stated once: Day Pass $0.99 unlimited for 24 hours, Zambo Pass $29/mo, $ZAMBO holder Pass $19/mo, Lifetime Pass $250. There is no annual tier.

Try it: Day Pass, $0.99 unlimited for 24 hours: https://zambo.dev/day-pass/ and add Zambo to your AI: https://zambo.dev/install/ (redirects to the add page, verified live today).

Sources

πŸ“° Read the original article on Dev.to AI

Originally published by Dev.to AI. Aggregated on AIWithGhost for educational purposes β€” full credit and traffic to the original publisher.