|
Hey there 👋,
Quick refresher in case you're newer here: Dograh is an open source voice AI platform - build, test, and run phone agents on the model and telephony stack of your choice.
Lots this month: a refreshed builder, text-based agent development, central telephony, plus MCP integrations, new realtime models, and a cleaner devcontainer setup.
|
A thank-you note first
We just crossed 3,000 stars on GitHub and 360 builders in the Slack community. Six months ago neither number had a comma in it. Thank you - every star, every issue, every Slack thread keeps us going.
→ Star us on GitHub
|
Refreshed agent builder - call directly from the canvas
The builder got a substantial visual refresh, and the highlight: you can now place test calls without leaving the page. No more jumping to a separate screen to dial in and check what you just changed - edit a node, hit Call, talk to the agent right there.
→ Open the builder
|
Develop your agent over text chat
You can now iterate on an agent over a text chat - type the customer side, see what the agent says back, then jump back to any previous turn to edit your user message or regenerate the agent response. The fastest way to debug a prompt without burning minutes on phone calls or TTS credits.
→ Try chat-based development
|
Multi-telephony configurations
A new telephony settings page where you register multiple providers and numbers in one place, then attach a specific number to a specific agent. Run inbound on Twilio and outbound on Plivo from the same workspace, or split traffic across numbers without rewiring anything.
→ Configure telephony
|
|
Also shipped this month
MCP integrations. Connect a Dograh agent to hundreds of services (Slack, Stripe, HubSpot, Linear, and more) by dropping in that service's MCP credentials. Tools show up in the agent's toolbox automatically.
Tuner integration. Drop a Tuner node into your workflow, paste three credentials, and every completed call streams to Tuner for transcript review, quality evals, and trend tracking - no custom observability code needed.
New realtime models. Ultravox, xAI Grok Voice, and OpenAI Realtime are now first-class speech-to-speech providers, joining Gemini 3.1 Live from last month.
Devcontainer setup. One-command local environment for contributors - open the repo in VS Code, accept the prompt, and the full stack comes up in a container. No more "works on my Mac".
|
|
How to access these
On app.dograh.com? You're already on the latest - nothing to do.
Self-hosted? Pull the new images and restart. The update guide covers both "latest" and pinned-tag flows.
→ Update your self-hosted stack
|
|
|
If you try any of this and something feels off - or if there's a model, provider, or integration you want next - hit reply. I read every message.
- Abhishek
Co-founder, Dograh
github.com/dograh-hq/dograh ★
|
|
