Mission
Telenumerus provides SMS verification infrastructure for developers and AI agents. We believe autonomous software should be able to access verification services programmatically, with the same reliability humans expect.
Every API endpoint, authentication method, and response format was designed with machine consumers in mind. No CAPTCHA gates, no manual approval, no friction between your code and a working phone number.
What we do
We aggregate phone numbers across multiple upstream providers and expose them through a single, consistent API. The system handles provider selection, failover, retry logic, and refunds automatically.
- 33+ supported platforms (Telegram, WhatsApp, Google, Discord, and more)
- Instant number delivery with automatic SMS polling
- Auto-retry across providers on failure
- Cryptocurrency payments (BTC, LTC, USDT, XMR)
- Webhook push delivery for incoming SMS
- AI-native API design: OpenAPI spec, llms.txt, AI plugin manifest, and a natural language ordering endpoint
Technical highlights
Every transaction is recorded in an immutable, hash-chained audit log. Each entry references the previous entry's hash, making retroactive tampering detectable.
Failed verifications and timed-out orders are refunded to your balance automatically. No support tickets, no waiting.
Five upstream providers with circuit breakers. If one provider fails or degrades, requests route to the next available source without interruption.
Server-Sent Events deliver SMS codes to your client the moment they arrive. No polling required.
For developers
The API is designed to be predictable and self-documenting.
- REST API with consistent JSON responses
- Two authentication methods: API key header or query parameter
- Structured error codes with human-readable messages
- Idempotency keys to safely retry requests
- Batch ordering (up to 20 numbers per request)
- Webhook delivery with HMAC signatures
- Full API documentation with request/response examples
For AI agents
Telenumerus is built to be consumed by autonomous software without human assistance.
- Self-service registration via API (no human approval needed)
- Natural language ordering endpoint for LLM-driven workflows
- Structured JSON responses parseable by any language model
- AI plugin manifest for ChatGPT and compatible agents
- OpenAPI 3.1 specification for automatic client generation
- llms.txt context file for LLM discovery