Service calls booked from live bays, not guesses
Dograh is the self-hostable voice agent platform with an open-source workflow builder. Answer service calls on your own numbers, book from live bay capacity, and run reminder and recall calls from your DMS while prices stay with your advisors.
One service call, from ring to booked bay
Check the owner first
Pre-Call Data Fetchfinds a likely record, proves nothing
The agent asks for the VIN's last four characters or the postcode before it names any vehicle.
→ Never read out a car before the check passes
Slots from live capacity
HTTP API toolreads your scheduler during the call
The agent offers two open slots for the job and writes the chosen one back to the schedule.
→ Send workflow_run_id so a retry never double-books
The visit, in the owner's words
Webhookthe outcome goes to your DMS
"A grinding noise when I brake" reaches the advisor as said, with the recording and transcript links.
→ Diagnosis waits for a technician
Reads from, writes to, runs on
Only after the owner check passes
Webhook node, retried automatically on a temporary failure
No built-in DMS integration, so nothing is locked to one vendor
A service line that stays on your side
Owner data stays home
Self-host Dograh and recordings sit on your servers. Audio stays inside your infrastructure only when the models run there too.
No pitch on a recall
Recall and reminder calls carry no offers, so a safety call never turns into a sales call.
A person when it matters
Price, warranty and diagnosis questions go to your service desk by transfer or a callback request.
No per-minute platform fee
Self-host the BSD-2 code and pay only for servers, phone minutes and the model keys you choose. Dograh Cloud has a 1¢-a-minute platform fee.
What changes when the platform is yours
| What you get | Dograh | hosted voice AI platforms |
|---|---|---|
| Open-source code you can read and change (BSD-2) | ✓ | ✕ |
| Self-host on your own servers or on-prem without an enterprise contract | ✓ | ✕ |
| No per-minute platform fee when you self-host | ✓ | ✕ |
| Run the whole stack, platform and models, on your own hardware | ✓ | ✕ |
Every dealership call, one deep dive each
Does the agent quote repair prices?+×
No. Prices, estimates and warranty decisions come from your service advisors. The agent records the question and hands it over.
How does it know which slots are free?+×
An HTTP API tool reads your scheduler or DMS during the call, so it offers only slots that exist right now.
Is there a built-in DMS integration?+×
No. Dograh connects through your DMS's own API using HTTP API tools, MCP tools, webhooks or tools like n8n.
Can it make recall calls?+×
Yes, as an extra prompt after the manufacturer's mailed notice. It reads the recall as written, books the free repair and makes no offer.
Do outbound reminder calls need consent?+×
Generally yes for AI-voice calls to mobiles under the TCPA, and adding an offer raises the bar to written consent. Confirm with counsel.
Does call audio stay on our servers?+×
Only when you self-host Dograh and run the speech and language models on your servers too. With hosted model keys, the audio reaches the provider.
Start with the inbound service line.
Self-host in one command, or start on Dograh Cloud.
$ curl … start_docker.sh && ./start_docker.sh