- DROP-OFFS
- FLOW DATA
- REVISIONS
- HUMAN
Short answer
How can I audit my onboarding flows using AI?
Your AI uses the Appcues connector to pull your current flow configurations and engagement data. It identifies specific steps where users stop interacting, giving you a clear map of where to fix your onboarding.
Audit outcomes.
Where your findings land.
The AI categorizes its findings into four distinct areas based on the data it pulls.
DROP-OFFS
Friction point identification
The AI flags specific steps in your flows where engagement spikes or falls. You get a list of exact moments users lose interest.
FLOW DATA
Current flow inventory
Your agent pulls a full list of active flows. This gives you a baseline of what is currently running in Appcues.
REVISIONS
Suggested flow tweaks
Based on the drop-off data, the AI suggests specific changes to your flow logic. You decide which changes to actually implement.
HUMAN
Manual design review
The AI identifies complex UI issues it can't see. These are the moments where you need to step in and look at the actual design.
The workflow
What your AI does when the request arrives.
The AI moves from broad discovery to specific data extraction to find your leaks.
Inventory flows
The AI first looks at everything you have running in Appcues to understand your current onboarding setup.
list_flowsInspect specific flows
Once it knows which flows exist, the AI pulls the granular details and engagement metrics for a specific flow you're worried about.
get_flowAnalyze drop-off points
The AI compares the steps within a flow to see where the user count drops significantly.
get_flowReport findings
The AI summarizes the data into a readable report of where your onboarding is failing.
list_flowsget_flow
Try it
Copy these to start.
Use these prompts to kick off your audit.
Starting points
These are opening lines. Replace the names of your flows with the actual names used in Appcues.
Appcues Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_rOwkjQxM3AikVsYjGJetdi9QldtT26neyniJs3YT/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Appcues capabilities are ready to use.
{
"mcpServers": {
"appcues-mcp": {
"url": "https://edge.vinkius.com/vk_preview_rOwkjQxM3AikVsYjGJetdi9QldtT26neyniJs3YT/mcp"
}
}
}List all my active Appcues flows so I can see what we're running.
Get the details for my 'New User Welcome' flow and tell me where people are dropping off.
Compare the engagement data between my 'Trial Onboarding' flow and my 'Pro Onboarding' flow.
Look at the 'Feature Tour' flow and identify the step with the highest exit rate.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Appcues once, then ask.
Just click the link to connect your account. Your credentials stay encrypted, and you can start chatting with your data immediately.
Connect Appcues to your AIFAQ
How this task behaves.
- 01
Can the AI change my Appcues flows directly?
No. The AI can only read your flow data and suggest changes. You must log in to Appcues to actually edit or publish any changes.
- 02
Can the AI delete a flow if it's performing poorly?
No. The AI has no permission to delete or modify your Appcues account settings or flows.
- 03
Does the AI see my user's personal information?
The AI only accesses the engagement and flow configuration data provided by the Appcues tools.
- 04
How does the AI know which flow to look at?
You tell the AI which flow to inspect by name, or you can ask it to list all flows first so you can pick one.
- 05
Can I use this with any AI client?
Yes, as long as your client is MCP-compatible, like Claude, Cursor, or Windsurf.
More questions about Appcues? See everything the Appcues Connector can do
Connect Appcues to Claude, Cursor, ChatGPT & more
