Start the Make Guided Connection path
Select Make in Clyvel Guided Connection to receive the gateway endpoint and the sequence for configuring the supported request path.
All integrations
MAKE INTEGRATION
Make AI Workflow Monitoring and OpenAI Gateway Integration
Scenario-based automation makes it easy to add AI calls quickly, but production teams still need to know what those calls cost, which provider handled them and where failures or latency originate. Clyvel adds an operational request layer around supported Make model traffic without replacing the automation platform.
WHEN THIS FITS
A practical production use case.
Make scenarios using supported OpenAI-compatible HTTP or model request patterns.
Teams that want AI spend and failures visible across multiple automations.
Organizations consolidating AI traffic from code and no-code tools into one operational view.
CONNECTION PATH
Connect the model request layer, then verify evidence.
Use the Clyvel Gateway for the model request
Update the relevant HTTP or OpenAI-compatible request so the scenario sends traffic through Clyvel with its scoped workload credential.
Run one scenario and verify evidence
Confirm the captured request in Clyvel, including provider, model, status, latency, tokens and estimated cost, before rolling the pattern out more broadly.
01
Central visibility for AI traffic created by automation scenarios.
02
Request-level cost and token evidence instead of invoice-only visibility.
03
Shared reliability context across Make and application code.
FAQ
Integration questions.
Can Make scenarios send AI traffic through Clyvel?
Clyvel includes a Guided Connection path for Make. The supported integration routes the model request through the Clyvel Gateway so operational telemetry can be captured.
Do I have to move the rest of my Make scenario into Clyvel?
No. Make remains responsible for the scenario. Clyvel is used for the production AI request layer around the supported model calls.
NEXT STEP
Connect one workload before expanding the rollout.
Start with a controlled request, confirm that operational evidence is trustworthy, and only then move more production traffic or add routing and reliability policy.