Automation · Customer service and scheduling
Tell a customer on WhatsApp whether their change is live, by checking GitLab
An AI agent finds the customer's request in GitLab, checks the merge request, its pipeline and the published release, and answers on WhatsApp.
How do I tell a customer on WhatsApp whether their change is published, by checking GitLab?
When a client of your agency asks on WhatsApp whether the change they requested is done, the Met finds the work item for that request in GitLab and checks its related merge requests. It opens the merge request to see whether it was merged and how its pipeline ended, and checks whether a published release came out afterwards. With that it answers in plain words whether the change is published, in review, or whether the deployment failed and the team is looking at it. At the end it leaves an internal comment on the work item so the team knows the client asked.
What starts it
When the customer writes
It starts when a client writes on WhatsApp to ask whether a change they requested is already on their site or app. There is no trigger to configure: the Met assigned to the line handles it in that conversation.
Nothing to set up. The Met assigned to that channel picks up the conversation and runs the procedure inside it.
Who runs it
Support Met for an agency that works in GitLab
Customer service team
This is not a named, seeded Met. It is the Met that answers the agency's or business's WhatsApp, with GitLab connected: its work ends when the client knows where their change stands and the team has the question noted.
Run by the customer support Met
Channels: WhatsApp
What you need connected
- GitLab Automate and collaborate
5 steps · 6 tools
The procedure, step by step
Each step shows the tool that runs. They belong to the real connector: if one of them stopped existing, this page would not build.
- 1
Find the client's request
The Met lists the work items in the client's project and searches by text for the request they mention, for example the contact form change. If there are several similar ones, it confirms with the client which one it is.
- 2
See which merge request solves it
The Met opens the work item and asks for its related merge requests. If there is none, it tells the client the request is logged and has no change in progress yet.
- 3
Check the merge request and its pipeline
The Met opens the merge request with its pipelines to learn whether it was merged, and opens the latest pipeline with its jobs to see whether it ended well or which job failed.
- 4
Confirm whether there is a published release
If the team publishes releases, the Met lists the project's releases and compares the date of the most recent one with the date the merge request was merged.
- 5
Answer and leave a record
The Met tells the client whether the change is published, in review or stopped by a failure, and leaves an internal comment on the work item so the team sees the client asked.
Before you start
What you need ready
The Met solves none of these for you. If one is missing, the automation stops there.
-
The client's requests as work items in GitLab
Your team logs each request as a work item, for example an issue, and links it to the merge request that solves it. Without that link, the Met cannot know which change matches.
-
GitLab connected in Met
In Met, open Connections → GitLab → Connect and sign in with an account that sees the client's project. Commenting arrives on; the admin decides whether to keep it that way.
-
A WhatsApp line with the Met assigned
The agency or business connects its WhatsApp Business API line and assigns the Met that serves clients. The question and the answer happen in that conversation.
The limits
What this automation does NOT solve
Better said here than discovered halfway through the rollout.
-
It does not publish or deploy the change. This procedure only reads the merge request, the pipeline and the releases; it does not run pipelines or merge.
-
It does not read release notes: GitLab lists releases without notes, so the Met decides by dates and says so when it cannot confirm.
-
It does not ask for confirmation in the chat before leaving the comment: if the action is on, it runs it. The admin decides in Connections which ones stay on.
-
It does not show the client the code, the log or the internal comments: it answers with the change's status in plain words.
FAQ
Questions about this automation
No. The client only talks to the Met on WhatsApp; everything read and commented in GitLab is done by the account the admin connected.
The Met tells the client the change is on hold and the team is reviewing it. If the owner asks, the Met can also read the log of the job that failed to explain what happened.
Yes. In that case the Met answers with what the merge request and the pipeline say: whether it was merged and whether the pipeline ended well, without naming a release.
Keep going
Other automations built on the same tools
Customer service and scheduling
Open a Linear issue from a customer report on WhatsApp and tell them how it is going
How do I make customer reports on WhatsApp arrive in Linear as issues?
Customer service
- Starts:
- when the customer writes
- Steps:
- 5, with 6 real tools
- Needs:
- Linear
Customer service and scheduling
Open a Freshdesk ticket from WhatsApp and tell where it stands
How do I turn a complaint that arrives on WhatsApp into a Freshdesk ticket without copying it by hand?
Customer service
- Starts:
- when the customer writes
- Steps:
- 5, with 10 real tools
- Needs:
- Freshdesk
Customer service and scheduling
Open a Gorgias ticket from WhatsApp and escalate it to the team that solves it
How do I move a complaint that arrives on WhatsApp into Gorgias without copying it by hand or losing what the customer said?
Customer service
- Starts:
- when the customer writes
- Steps:
- 5, with 7 real tools
- Needs:
- Gorgias
Customer service and scheduling
Open a Jira issue from a report that comes in on WhatsApp and answer how it is going
How do I make customer reports on WhatsApp arrive in Jira as issues?
Customer service
- Starts:
- when the customer writes
- Steps:
- 5, with 6 real tools
- Needs:
- Jira y Confluence
Want this automation running on your own data?
We will show it to you with your accounts connected, not with a canned demo.