outbound-callMake outbound phone calls via ElevenLabs voice agent and Twilio
Install via ClawdBot CLI:
clawdbot install humanjesse/outbound-callSource code and setup guide: github.com/humanjesse/hostinger-openclaw-guides
Place outbound phone calls using the ElevenLabs voice agent with Twilio. The voice agent on the call uses OpenClaw as its brain — same as inbound calls.
When the user asks you to:
Run the call script with a phone number in E.164 format:
python3 skills/outbound-call/call.py +1XXXXXXXXXXWith an optional custom first message (what the agent says when the recipient picks up):
python3 skills/outbound-call/call.py +1XXXXXXXXXX "Hi John, I'm calling about your appointment tomorrow."With optional call context (passed as a dynamic variable to the agent):
python3 skills/outbound-call/call.py +1XXXXXXXXXX "Hi, this is a quick follow-up call." "Customer requested callback about billing issue #4521"Generated Mar 1, 2026
Automated outbound calls to remind patients or clients about upcoming appointments. The agent can confirm attendance, provide details, and handle rescheduling requests, reducing no-shows and administrative workload.
Call customers after a support ticket is resolved to ensure satisfaction and gather feedback. The agent can reference specific issue details from the call context, offering personalized and efficient service.
Proactively call leads to introduce products or services, qualify interest, and schedule demos. The agent uses a custom first message to engage recipients and can adapt the conversation based on responses.
Place urgent calls to notify employees or community members about critical updates, such as office closures or safety alerts. The agent delivers pre-recorded messages quickly and can handle basic inquiries.
Call customers with overdue invoices to remind them of payments, explain charges, and assist with payment methods. The agent can reference account details from the call context to provide accurate information.
Offer the outbound call skill as part of a subscription-based platform for businesses, charging monthly fees based on call volume or features. Revenue comes from recurring subscriptions and tiered pricing plans.
Charge customers per call made using the skill, with pricing based on call duration or number of recipients. This model suits businesses with sporadic or low-volume calling needs, providing flexibility and cost control.
License the skill to other companies or developers who integrate it into their own products under their brand. Revenue is generated through upfront licensing fees or ongoing royalties based on usage.
💬 Integration Tip
Ensure all required environment variables (ELEVENLABS_API_KEY, ELEVENLABS_AGENT_ID, ELEVENLABS_PHONE_NUMBER_ID) are securely configured before use, and test calls in a controlled environment to verify functionality.
Transcribe audio via OpenAI Audio Transcriptions API (Whisper).
Local speech-to-text with the Whisper CLI (no API key).
ElevenLabs text-to-speech with mac-style say UX.
Text-to-speech conversion using node-edge-tts npm package for generating audio from text. Supports multiple voices, languages, speed adjustment, pitch control, and subtitle generation. Use when: (1) User requests audio/voice output with the "tts" trigger or keyword. (2) Content needs to be spoken rather than read (multitasking, accessibility, driving, cooking). (3) User wants a specific voice, speed, pitch, or format for TTS output.
End-to-end encrypted agent-to-agent private messaging via Moltbook dead drops. Use when agents need to communicate privately, exchange secrets, or coordinate without human visibility.
Text-to-speech via OpenAI Audio Speech API.