Handover from Goose to dots (ChatGPT)

A job started in Goose rarely has to finish there. A Zambo HANDOVER moves the job itself to dots (ChatGPT) as one checkable record: the goal, the finished steps with a verifiable receipt for each, the decisions, and the exact next step. The receiver in dots (ChatGPT) starts from the record, not from a summary of it. HANDOVER is Zambo's flagship product.

Before you start: Zambo on the Goose side

Run goose configure, choose Add Extension, then Remote Extension (Streamable HTTP), name it zambo, and enter https://zambo.dev/mcp as the endpoint URI. Or add a zambo entry under extensions in ~/.config/goose/config.yaml with type streamable_http, uri set to https://zambo.dev/mcp, enabled true, and timeout 300. The type must be exactly streamable_http and the address must sit under uri: a url field will not load. Extensions apply to new sessions, so start a fresh session after configuring.

Full setup: Add Zambo to Goose. The free tier is 20 free calls per day. Work run through Zambo in Goose saves a verifiable receipt per call as it runs, and those receipts are what the handover carries.

Step 1: finish the Goose side and create the handover

  1. Do the work in Goose through Zambo tools, so each finished step has its receipt saved at the moment it ran.
  2. At the end of the Goose session, give one wrap up instruction: hand this job over with the goal, what is done, what is next, every decision, and the receipts attached. Zambo assembles the handover as a single record and returns a link in the pattern https://zambo.dev/handover/ followed by the handover ID. Send that one link to whoever continues the job in dots (ChatGPT). Nothing else needs to be copied across: no recap, no pasted chat history.
  3. Open the returned handover link yourself before sending it on. Check the goal reads the way the job was commissioned, the next step is exact, and any step nobody verified is named as not checked.

Step 2: connect Zambo on the dots (ChatGPT) side

A dot uses the plugins you have already set up in ChatGPT. Add Zambo in ChatGPT on the web first (Plugins, plus button, Add custom MCP server, name Zambo, server URL https://zambo.dev/mcp, create the plugin), then install the plugin. OpenAI says plugin permissions are shared across dots, ChatGPT, ChatGPT Work, and Codex, so the installed connection is what your dot can use, within the permissions you granted.

Full setup: Add Zambo to dots (ChatGPT). New to Zambo on either side? The bridge page walks the whole install: Add Zambo to your AI.

Step 3: open the handover in dots (ChatGPT) and continue

  1. Open the handover link in your browser first and read the record: the goal, the finished steps with their receipts, the decisions, and the exact next step.
  2. Confirm the Zambo plugin is installed in ChatGPT, not just created, and that its permissions are enabled in the Plugins tab. A dot can only use plugins that are already set up.
  3. Give your dot the handover link and say: open this handover, check the receipts that carry weight, record accept or flag on the handover, then continue from the recorded next step. Note the documented limit: during proactive background research a dot uses connected tools read only, so actions that change things still go through the usual approvals.

What travels in the handover

Verify note

Before accepting, open one or two receipt links from inside the handover and paste each receipt ID into zambo.dev/verify. A match confirms the saved result was not changed since it was saved. That is what a receipt proves: the record is intact, not that the work was right. Judging the work is the receiver's job, now done against the record instead of a recap. Then record accept, or flag what is wrong, on the handover itself. The transfer is not complete until that decision exists.

Why not just paste a summary into dots (ChatGPT)?

A summary is narration: it can skip the step that failed and drop the decision that constrained everything after it, and none of that is detectable later. The handover record fixes each step under its receipt at the time it ran in Goose, so dots (ChatGPT) can recompute any step instead of believing it. The general pattern is on the flagship page: AI agent handover.

More handover recipes

A handover is only as strong as the receipt the receiving side can check. The AER-1 badge kit carries that same verifiable-receipt proof outward, as an embeddable badge for the pages and profiles where the work is shown.