Automation · Operations and reporting
Ask on WhatsApp for a change to an app or site built with Lovable and publish it
An AI agent takes the change your client asks for on WhatsApp, hands it to the Lovable agent, shows you what changed and publishes the app.
How do I let my clients ask for changes to their Lovable app on WhatsApp without losing control?
A Met connected to Lovable takes the change your agency's client asks for on WhatsApp, such as new opening hours or a homepage line. It finds their project, passes the request to the Lovable agent and waits for the answer. Then it pulls the diff, sums it up in plain words and shares the preview so someone can approve it. Once you give the go-ahead, it publishes the app and sends back the live URL. Each request to the agent spends credits from your Lovable plan, and publishing only happens if the admin left that action switched on.
What starts it
When the customer writes
It starts when an agency client writes on WhatsApp to ask for a change to their app or site. There is no trigger to configure: the Met assigned to the line runs the procedure inside 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 builds apps on Lovable
Customer service team
This is not a named, seeded Met. It is the Met that answers the agency's WhatsApp, with Lovable connected. Its work ends when the change the client asked for is done and published, or when it sits in the preview waiting for someone's approval.
Run by the customer support Met
Channels: WhatsApp
What you need connected
- Lovable Automate and collaborate
6 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 project
The Met searches the workspace's projects for the one that belongs to the client who is writing, by its name. If two look similar, it asks which one.
- 2
Hand the change to the Lovable agent
The Met sends the project's agent the change exactly as the client asked for it and waits for the answer. This message spends credits from the agency's Lovable plan.
- 3
Review what changed
Once the work is done, the Met pulls that message's diff and sums it up in plain words, so nobody has to read code to know what was touched.
- 4
Share the preview
The Met pulls the project details, with the preview URL and a screenshot, and sends it to the client or to the agency person who approves changes.
- 5
Publish once it is approved
With the go-ahead, the Met publishes the app and sends the live URL. If the security scan found unresolved critical findings, Lovable refuses to publish and the Met says so.
- 6
Keep the request in the history
If the client later asks what was changed and when, the Met checks the project's edit history and answers with what was recorded.
Before you start
What you need ready
The Met solves none of these for you. If one is missing, the automation stops there.
-
Lovable connected in Met
In Met, open Connections → Lovable → Connect with the agency's account. The admin leaves sending messages to the agent on and, if they want the Met to publish, publishing too.
-
Credits on the Lovable plan
Each request to the agent spends Lovable credits. Check the workspace balance and the plan before opening this channel to all your clients.
-
A rule for who approves
Write in the Met's skill who gives the go-ahead before publishing: the client or someone at the agency. Without that rule, the Met does not know whom to wait for.
The limits
What this automation does NOT solve
Better said here than discovered halfway through the rollout.
-
It does not judge whether the change came out right: the Lovable agent makes it and the Met sums it up. A person does the final review in the preview.
-
It does not ask for confirmation in the chat before sending the request or publishing: if the action is on, it runs it. Leave publishing off if you want to do it yourself.
-
It does not cap spending: each request to the agent spends credits, and a client who asks for many changes in a row spends them all the same.
-
It does not publish if the security scan found unresolved critical findings: Lovable refuses it and someone has to review them in the project.
FAQ
Questions about this automation
They can ask, but the skill should say which kinds of changes the Met accepts, such as copy, hours or prices, and which ones go to a person at the agency.
It gets fixed before publishing, with another request to the agent or from the Lovable editor. That is why reviewing the diff and the preview comes before publishing.
Only if the admin left the publish action on. If you would rather review first, leave it off in Connections and publish yourself from Lovable.
Whatever Lovable charges for each message to the project's agent. Pulling the diff, the preview or the history does not spend credits.
Keep going
Other automations built on the same tools
Operations and reporting
Turn off on WhatsApp the feature flag that breaks the checkout in LaunchDarkly
How do I quickly turn off a feature flag that broke the checkout without opening LaunchDarkly?
- Starts:
- when someone asks for it
- Steps:
- 5, with 7 real tools
- Needs:
- LaunchDarkly
Operations and reporting
Build a department's higher education report without opening SNIES
How do I pull SNIES enrollment and program figures without going into the portal?
- Starts:
- when the customer writes
- Steps:
- 5, with 5 real tools
- Needs:
- SNIES
Operations and reporting
Build your team's org chart in Lucid on WhatsApp and share it
How do I build my business's org chart without sitting down to draw it?
- Starts:
- when someone asks for it
- Steps:
- 6, with 6 real tools
- Needs:
- Lucid
Operations and reporting
Tell customers on your status page that the store is down, asking on WhatsApp with Better Stack
How do I tell my customers my online store is down without opening Better Stack?
- Starts:
- when someone asks for it
- Steps:
- 6, with 6 real tools
- Needs:
- Better Stack
Want this automation running on your own data?
We will show it to you with your accounts connected, not with a canned demo.