phone-calls-blandMake AI-powered phone calls via Bland AI - book restaurants, make appointments, inquire about services. The AI calls on your behalf and reports back with transcripts.
Install via ClawdBot CLI:
clawdbot install dru-ca/phone-calls-blandMake AI-powered phone calls on behalf of the user — book restaurants, make appointments, inquire about services, etc.
Why Bland AI?
phone_number + task to make a callAdd to your environment or .env:
BLAND_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxx
Or store in ~/.clawd/secrets.json:
{
"bland_api_key": "sk-xxxxxxxxxxxxxxxxxxxxxxxx"
}
./phone-call.sh "+447123456789" "Call the restaurant and book a table for 2 at 7pm tonight under the name John"
./phone-call.sh "+447123456789" "Ask about their opening hours" --voice maya
./check-call.sh <call_id>
Restaurant Booking:
Call this restaurant and book a table for 4 people on Saturday at 8pm.
The booking should be under the name "Smith". If they ask for a phone
number, give them +447123456789.
Appointment Inquiry:
Call this dental office and ask what appointments are available next
week for a routine checkup. Get at least 3 options if possible.
Service Inquiry:
Call this plumber and ask if they can come out tomorrow to fix a
leaking tap. Get a quote for the callout fee.
Built-in voices:
josh - Male, professionalmaya - Female, friendly (default)florian - Male, European accentderek - Male, casualjune - Female, professionalnat - Male, naturalpaige - Female, upbeatThe AI can detect voicemails and either hang up, leave a message, or ignore.
Set record: true to get a recording URL after the call.
Get notified when calls complete by setting a webhook URL.
For complex flows (IVR menus, multi-step processes), create pathways in the Bland dashboard.
"Invalid phone number"
+44 for UK, +1 for US"Insufficient balance"
"Rate limit exceeded"
phone-call.sh — Make a phone callcheck-call.sh — Check call status/transcriptbland.sh — Low-level API wrapperGenerated Mar 1, 2026
Automate table bookings by having the AI call restaurants to reserve seats for customers. It handles details like party size, time, and name, freeing staff from phone tasks. Ideal for busy dining establishments or concierge services.
Use the AI to call clinics or dental offices to inquire about available slots and book appointments for patients. It can gather multiple options and confirm details, reducing administrative workload in healthcare settings.
Automate calls to plumbers, electricians, or other service providers to request quotes for repairs or installations. The AI can describe issues and schedule callouts, streamlining procurement for property management.
Implement AI calls to follow up with customers after purchases or support tickets, gathering feedback or confirming satisfaction. This enhances engagement without manual effort in retail or tech support.
Deploy the AI to call event attendees to confirm details, send reminders, or handle last-minute changes. Useful for organizers in entertainment or corporate sectors to improve attendance rates.
Charge customers based on call minutes used, leveraging Bland AI's per-minute pricing. This model suits agencies or freelancers offering on-demand calling services with transparent billing and low upfront costs.
Offer monthly plans for businesses to automate routine calls like appointment bookings or inquiries. Include a set number of minutes per month, providing predictable revenue and scaling with client needs.
Integrate the skill into existing platforms (e.g., CRM or scheduling software) and license it to other companies. This generates revenue through licensing fees or commissions on calls made via the integrated system.
💬 Integration Tip
Store the BLAND_API_KEY securely in environment variables or a secrets file to avoid hardcoding, and test with simple tasks first to ensure proper setup.
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.