dograh-hq/dograh

Open source voice AI platform. Self-hosted alternative to Vapi and Retell. On Prem, BYOK across Speech to Speech or LLM/STT/TTS, with a visual workflow builder, MCP native and telephony support.

Dograh AI – Open‑source platform for building voice agents

Dograh is a self‑hostable, fully open‑source alternative to commercial voice‑AI services such as Vapi and Retell. It lets you create, test, and run conversational voice bots through a visual workflow editor, while giving you complete control over the underlying LLM, speech‑to‑text (STT), text‑to‑speech (TTS), and telephony providers.

What it does

  • Visual workflow builder – drag‑and‑drop nodes (start, agent, tools, QA, webhooks, end) to define call flows.
  • Instant testing – test bots in the browser with live audio or a fast “Test Chat” text interface that can replay and edit turns.
  • Telephony integration – built‑in connectors for Twilio, Vonage, Telnyx, Plivo, Vobiz, Cloudonix, Asterisk ARI, plus human‑hand‑off via call transfer.
  • Pluggable AI stack – you can use Dograh’s bundled LLM/STT/TTS stack or plug in any provider you prefer (OpenAI, Anthropic, etc.).
  • Developer‑friendly SDKs – Python (dograh-sdk) and Node (@dograh/sdk) libraries for programmatic creation of agents and outbound calls.
  • MCP (Machine‑Code‑Programming) integration – connect coding assistants like Claude Code, Codex, or Cursor to the platform so they can read/write workflows, fetch schemas, and edit agents directly from the IDE.

Why it matters

  • Open source & BSD‑2 license – no vendor lock‑in; you can inspect, modify, and self‑host the entire stack.
  • Self‑hosting in one command – a single Docker‑compose pull‑and‑run script gets the whole system up in ~2 minutes.
  • Data residency & privacy – all data stays on your infrastructure (or on Dograh’s managed cloud if you prefer).
  • Flexible pricing – free when self‑hosted; the cloud version is usage‑based.

Quick start (local)

# Grab the Docker compose file and starter script
curl -o docker-compose.yaml \
  https://raw.githubusercontent.com/dograh-hq/dograh/main/docker-compose.yaml \
 && curl -o start_docker.sh \
  https://raw.githubusercontent.com/dograh-hq/dograh/main/scripts/start_docker.sh \
 && chmod +x start_docker.sh \
 && ./start_docker.sh   # pulls images and starts services

After the containers are healthy, open http://localhost:3010. The UI walks you through creating your first voice bot (choose inbound/outbound, give a short description, then hit Test Agent). You can talk to the bot via the browser’s audio test or switch to the text‑chat tester for rapid iteration.

Using an AI coding assistant to set up Dograh

If you have Claude Code, Codex, or another MCP‑compatible assistant, you can install the official Dograh plugin (dograh-hq/dograh-plugins) and then ask the assistant to “set up Dograh” – it will run the same Docker script, verify the deployment, and even help you configure keys.

Extending the platform

  • Add your own LLM / TTS / STT – configure provider keys in the UI or via environment variables; Dograh’s modular backend will route calls accordingly.
  • Custom nodes & tools – define webhooks, knowledge‑base lookups, or arbitrary tool calls inside a workflow.
  • Programmatic control – use the Python or Node SDKs to create agents, trigger outbound calls, or embed Dograh functionality in your own applications.

Community & support

  • Slack community – active channel for questions, feature discussion, and onboarding.
  • GitHub Discussions & Issues – place use‑case requests, bug reports, or PRs.
  • Documentation – full docs at https://docs.dograh.com covering deployment, MCP integration, and troubleshooting.

License

BSD 2‑Clause – permissive, allowing commercial use, modification, and redistribution.

In a nutshell: Dograh AI is a production‑grade, open‑source voice‑assistant platform that combines a visual workflow editor, telephony connectors, and a pluggable AI stack, all runnable with a single Docker command. It targets teams that need full control over their voice AI without being locked into a SaaS provider.

Related

  • Project
    EpicStaff/EpicStaff

    EpicStaff is a self‑hosted, Django‑backed platform that lets engineers design multi‑agent AI workflows with a drag‑and‑drop visual editor, while operations teams can own, audit, and modify those flows without writing code. It supports LLMs, RAG, custom Python tools, persistent memory, webhooks, and enterprise‑grade security, all runnable via Docker.

  • Project
    labring/FastGPT

    FastGPT is an open‑source platform for building AI agents and LLM‑driven applications. It provides a visual Flow editor, knowledge‑base (RAG) support, plugin hot‑reloading, debugging tools, and can be run locally via Docker or deployed on Sealos/Kubernetes. A hosted SaaS version and a commercial edition are also offered.

  • Project
    theopenconversationkit/tock

    Tock is an open‑source conversational AI platform that provides an NLP stack, a visual studio for designing dialogue flows, a Kotlin‑based DSL (with Python, Node.js and REST bindings), ready‑made connectors to chat/voice channels, and Docker‑based deployment.

  • Project
    Paca-AI/paca

    An open-source, self-hosted project management platform where AI agents and humans collaborate as equal teammates within a Scrum team.

  • Project
    bolna-ai/bolna

    Bol na is an open‑source framework that orchestrates speech‑to‑text, LLM, and text‑to‑speech services to build voice‑first assistants that can place and receive phone calls. It ships Docker‑compose containers for the core server, telephony webhook (Twilio/Plivo), Redis, and ngrok, and provides a Python SDK (`Assistant`) for building streaming pipelines. The platform is provider‑agnostic (supports Deepgram, Azure, OpenAI, ElevenLabs, AWS Polly, etc.) and is extensible to new telephony or audio services.