How do I install the Openclaw Voice Gpt Realtime skill?

Run npx skillmds@latest add modbender/openclaw-voice-gpt-realtime in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Openclaw Voice Gpt Realtime safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Openclaw Voice Gpt Realtime?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Openclaw Voice Gpt Realtime free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Openclaw Voice Gpt Realtime?

modbender (@modbender) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Openclaw Voice Gpt Realtime by modbender

Name: Openclaw Voice Gpt Realtime
Rating: 4.2 (5 reviews)
Author: modbender

File contents

Voice Calls (OpenAI Realtime)

Make real phone calls through your OpenClaw agent. Ask it to book a restaurant, check store hours, schedule an appointment — it dials the number, handles the conversation, and reports back with structured results.

Uses OpenAI's Realtime API for single-model speech-to-speech with ~200-300ms response latency. No separate STT or TTS — one model does it all.

Setup

This skill requires a Twilio account and an OpenAI API key with Realtime API access.

Set your credentials in the plugin config (via OpenClaw settings or openclaw.json):
- twilio.accountSid — your Twilio Account SID
- twilio.authToken — your Twilio Auth Token
- fromNumber — a Twilio voice-capable phone number (E.164 format, e.g. +17075551234)
- openai.apiKey — your OpenAI API key
- publicUrl — a public HTTPS origin that routes to the plugin's server (port 3335 by default). Must not be localhost/private/internal.
Set up a tunnel (Cloudflare Tunnel, ngrok, Tailscale Funnel, etc.) so Twilio can reach the webhook server.
Verify setup:

openclaw voicecall-rt status

Usage

Just tell your agent what to call and why:

"Call Tony's Pizza at +14155551234 and reserve a table for 4 on Friday at 7pm"

"Call the barbershop at +14155559876 and book a haircut for Saturday morning"

"Call +14155550000 and ask if they have the iPhone 16 Pro in stock"

The agent writes a system prompt for the voice AI, dials the number, and the voice AI handles the conversation autonomously — including navigating phone menus (DTMF), detecting voicemail, and reporting the outcome. The plugin wraps prompts with safety guardrails and blocks deceptive identity behavior.

CLI

openclaw voicecall-rt call -n +14155551234 -t "Check store hours"
openclaw voicecall-rt status
openclaw voicecall-rt active

Inbound calls

Optionally receive calls by enabling inbound.enabled and setting a policy (open or allowlist). Disabled by default.

Cost

~~$0.31/min total (~~$0.06 OpenAI input + ~$0.24 OpenAI output + ~$0.014 Twilio). A typical 5-minute call costs ~$1.55.

Notes

The voice AI waits for the callee to speak before talking ("listen first") — no awkward overlap on pickup.
Server binds to 127.0.0.1 by default. Only exposed via your tunnel.
Max 5 concurrent calls by default (configurable via calls.maxConcurrent).
Debug mode (debug: true) enables call recording, verbose logging, and latency metrics; recordings/transcripts may contain sensitive data.

---name: openclaw-voice-gpt-realtimedescription: Make real phone calls through your OpenClaw agent via OpenAI's Realtime API. ~200-300ms latency, natural voice, IVR navigation, voicemail detection.--- # Voice Calls (OpenAI Realtime) Make real phone calls through your OpenClaw agent. Ask it to book a restaurant, check store hours, schedule an appointment — it dials the number, handles the conversation, and reports back with structured results. Uses OpenAI's Realtime API for single-model speech-to-speech with ~200-300ms response latency. No separate STT or TTS — one model does it all. ## Setup This skill requires a Twilio account and an OpenAI API key with Realtime API access. 1. Set your credentials in the plugin config (via OpenClaw settings or `openclaw.json`): - `twilio.accountSid` — your Twilio Account SID - `twilio.authToken` — your Twilio Auth Token - `fromNumber` — a Twilio voice-capable phone number (E.164 format, e.g. `+17075551234`) - `openai.apiKey` — your OpenAI API key - `publicUrl` — a public HTTPS origin that routes to the plugin's server (port 3335 by default). Must not be localhost/private/internal. 2. Set up a tunnel (Cloudflare Tunnel, ngrok, Tailscale Funnel, etc.) so Twilio can reach the webhook server. 3. Verify setup: ```bash openclaw voicecall-rt status ``` ## Usage Just tell your agent what to call and why: > "Call Tony's Pizza at +14155551234 and reserve a table for 4 on Friday at 7pm" > "Call the barbershop at +14155559876 and book a haircut for Saturday morning" > "Call +14155550000 and ask if they have the iPhone 16 Pro in stock" The agent writes a system prompt for the voice AI, dials the number, and the voice AI handles the conversation autonomously — including navigating phone menus (DTMF), detecting voicemail, and reporting the outcome. The plugin wraps prompts with safety guardrails and blocks deceptive identity behavior. ### CLI ```bash openclaw voicecall-rt call -n +14155551234 -t "Check store hours" openclaw voicecall-rt status openclaw voicecall-rt active ``` ### Inbound calls Optionally receive calls by enabling `inbound.enabled` and setting a policy (`open` or `allowlist`). Disabled by default. ## Cost ~$0.31/min total (~$0.06 OpenAI input + ~$0.24 OpenAI output + ~$0.014 Twilio). A typical 5-minute call costs ~$1.55. ## Notes - The voice AI waits for the callee to speak before talking ("listen first") — no awkward overlap on pickup. - Server binds to `127.0.0.1` by default. Only exposed via your tunnel. - Max 5 concurrent calls by default (configurable via `calls.maxConcurrent`). - Debug mode (`debug: true`) enables call recording, verbose logging, and latency metrics; recordings/transcripts may contain sensitive data.

Frequently asked questions

Openclaw Voice Gpt Realtime