The Scenario
Your team just finished onboarding two new services into the federated graph. That brings the total to eleven subgraphs across three branches, and nobody has a current list of what is deployed where.
The platform engineering manager asked for a documentation workbook before end of day — subgraph names, endpoint URLs, branch assignments, and schema sizes — so the team can review it before the next federation meeting.
The bad version:
- Log into the Grafbase dashboard, click into each subgraph one by one, and manually copy the name, endpoint URL, and branch into a row in your Excel workbook
- Switch back to the SDL view for each subgraph, count the type definitions by eye or copy the full SDL text and paste it into a word counter somewhere
- Repeat eleven times, then go back and verify you did not accidentally mix up two endpoints that share a similar naming pattern
Eleven subgraphs sounds manageable until you realize you missed one, the branch label format changed mid-way through your copy session, and the SDL character counts are already wrong because a colleague deployed a patch while you were working.
Nobody hired you to do registry transcription. Your job is to understand what is in the graph, not to manually mirror what Grafbase already knows.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that runs inside your Excel workbook. It reads the workbook, understands the context, and connects to Grafbase to pull registry data on your behalf. Open the SheetXAI sidebar, and ask.
List all published Grafbase subgraphs for account my-org and write each subgraph's name, endpoint URL, and branch into columns A–C of Sheet1
What You Get
- Column A: subgraph name, one per row
- Column B: the endpoint URL for each subgraph
- Column C: the branch it is published on
- If you asked for schema size: character count in column D, number of type definitions in column E
SheetXAI surfaces errors in a note on the affected row — if a subgraph endpoint returns a connection error or the SDL fetch fails, you see that inline rather than discovering a silent gap later.
What If the Data Is Not Quite Ready
The endpoint URLs are in the workbook but the branch column is blank for three rows
For rows in Sheet1 where column C is empty, look up the branch for each subgraph name in column A from Grafbase and fill it in
The subgraph names in the workbook use a different casing convention than Grafbase returns
Fetch all Grafbase subgraphs for account my-org, normalize each name to lowercase-with-hyphens, and write them into column A of Sheet1 — write the Grafbase-original name into column F for reference
You need to join subgraph data with a second worksheet that tracks ownership by team
For each subgraph in Sheet1 column A, look up the owning team name from the Team column in Sheet2 (matched by subgraph name) and write the team name into column D of Sheet1
You need cleanup, schema sizing, and a summary written in one shot
Pull all Grafbase subgraphs for account my-org, normalize names to lowercase, write name/endpoint/branch/character-count into columns A–D of Sheet1, then write a summary row in Sheet2 showing total subgraph count, average schema size, and which branch has the most subgraphs
One prompt handles the registry pull, the normalization, the per-subgraph sizing, and the cross-worksheet summary. The pattern is to ask for the cleanup and the deliverable together — not in separate rounds.
Try It
Open an Excel workbook you use to track your GraphQL infrastructure, then Get the 7-day free trial of SheetXAI and ask it to pull your Grafbase subgraph registry into the workbook. Also worth reading: Audit All Grafbase API Keys Into an Excel workbook and the Grafbase overview for Excel.