Docker

docker compose up

See Dockerfile and docker-compose.yml for the default configuration.

Quickstart

Set the env vars your carrier and engine need:

Twilio

export TWILIO_ACCOUNT_SID=ACxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
export TWILIO_AUTH_TOKEN=your_auth_token
export OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxx

Telnyx

export TELNYX_API_KEY=KEYxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
export TELNYX_CONNECTION_ID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
export OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxx

<details open> <summary><strong>Python</strong></summary>

pip install getpatter

from getpatter import Patter, Twilio, OpenAIRealtime

phone = Patter(carrier=Twilio(), phone_number="+15550001234")
agent = phone.agent(engine=OpenAIRealtime(), system_prompt="You are a friendly receptionist for Acme Corp.", first_message="Hello! How can I help?")
await phone.serve(agent, tunnel=True)

Or with Telnyx:

from getpatter import Patter, Telnyx, OpenAIRealtime

phone = Patter(carrier=Telnyx(), phone_number="+15550001234")
agent = phone.agent(engine=OpenAIRealtime(), system_prompt="You are a friendly receptionist for Acme Corp.", first_message="Hello! How can I help?")
await phone.serve(agent, tunnel=True)

</details>

<details> <summary><strong>TypeScript</strong></summary>

npm install getpatter

import { Patter, Twilio, OpenAIRealtime } from "getpatter";

const phone = new Patter({ carrier: new Twilio(), phoneNumber: "+15550001234" });
const agent = phone.agent({ engine: new OpenAIRealtime(), systemPrompt: "You are a friendly receptionist for Acme Corp.", firstMessage: "Hello! How can I help?" });
await phone.serve({ agent, tunnel: true });

Or with Telnyx:

import { Patter, Telnyx, OpenAIRealtime } from "getpatter";

const phone = new Patter({ carrier: new Telnyx(), phoneNumber: "+15550001234" });
const agent = phone.agent({ engine: new OpenAIRealtime(), systemPrompt: "You are a friendly receptionist for Acme Corp.", firstMessage: "Hello! How can I help?" });
await phone.serve({ agent, tunnel: true });

</details>

tunnel: true spawns a Cloudflare quick tunnel and points your Twilio number at it — great for dev / acceptance. For production outbound calls (especially on Twilio), replace it with ngrok or a static webhook_url to avoid WSS upgrade races on first call. See Tunneling for details.

Every carrier and provider reads its credentials from environment variables by default; see each SDK's README for the full catalog.

Example: clone and run the inbound agent template

git clone https://github.com/PatterAI/patter-inbound-agent cd patter-inbound-agent cp .env.example .env # fill in your keys cd python && pip install -r requirements.txt && python main.py ```

Configuration

Environment Variables

```bash cp .env.example .env

Edit .env with your API keys

```

Telnyx: Telnyx is a fully supported telephony provider alternative to Twilio. Both carriers receive equal support for DTMF, transfer, and metrics. Recording parity is supported via Telnyx Call Control; consult the Telnyx portal for configuration details.

API Reference

Python SDK

cd libraries/python && pip install -e ".[dev]" && pytest tests/ -v

TypeScript SDK

cd libraries/typescript && npm install && npm test ```

Please open an issue before submitting large changes so we can discuss the approach first.

How Patter compares

Patter is purpose-built for production voice over real telephony. Out of the box you get Twilio + Telnyx parity (DTMF, transfer, AMD, voicemail drop, recording), both architectures from one API — speech-to-speech (Realtime / ConvAI engines) and the sandwich pipeline (STT → LLM → TTS) — and production-grade barge-in / VAD / IVR primitives that work the same on every carrier. Observability is vendor-neutral OpenTelemetry tracing, plus a built-in dashboard and tunnel; no extra collector required. The 4-line quickstart above replaces ~50 lines of glue you'd otherwise write against a generic voice-agent toolkit, and the Python and TypeScript SDKs are identical — same surface, same hooks, same events — so cross-runtime teams ship the same agent twice without rewriting it.