codasms-mcp
Enables AI agents to purchase phone numbers and retrieve one-time SMS/OTP verification codes via the CODASMS API, with support for 700+ services and 180+ countries including automatic refunds if no code arrives.
README
<div align="center">
CODASMS MCP Server
Receive one-time SMS/OTP verification codes from your AI agent. An MCP server over the CODASMS Reseller API — get a number, wait for the code, release it — across 700+ services and 180+ countries, pay only on delivery.
</div>
This server lets an AI agent (Claude Desktop, Cursor, or anything that speaks MCP) buy a phone number and read the verification code programmatically. It's a thin wrapper over the public CODASMS Reseller API — it holds no logic of its own; every charge and refund is decided server-side by the API.
You bring your own key. Getting one is free to sign up: https://codasms.com/reseller.
Tools
| Tool | What it does |
|---|---|
get_balance |
Your current spendable balance. |
list_services |
Buyable services with live price + stock (filter by service / country). |
list_countries |
Countries with buyable numbers (optional service filter). |
get_number |
Buy a number for a service+country. Charges your balance (auto-refunds if no code in 20 min). |
wait_for_otp |
Poll an order until the code arrives (or timeout / refund). |
check_order |
One-shot status check for an order. |
release |
Cancel an order and refund now (unless the code already arrived). |
get_number accepts an optional max_price_cents (or set CODASMS_MAX_PRICE_CENTS) so an
autonomous agent can't buy a number priced above your cap.
Quick start (local / stdio)
export CODASMS_API_KEY=coda_live_xxxxxxxx # from https://codasms.com/reseller
npx codasms-mcp
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"codasms": {
"command": "npx",
"args": ["-y", "codasms-mcp"],
"env": { "CODASMS_API_KEY": "coda_live_xxxxxxxx" }
}
}
}
Then ask: "Get me a Telegram number in the UK and read the code."
Remote (HTTP) mode
The same tools are served over Streamable HTTP, so you can host one instance for many users — each sends their own key as a Bearer token (nothing is stored server-side):
npm run build && npm run start:http # listens on :8080, POST /mcp
POST /mcp
Authorization: Bearer coda_live_xxxxxxxx
Content-Type: application/json
src/http.ts is a portable Node server (Deno Deploy, Railway, Fly, any container). For
Cloudflare Workers, swap the Node http handler for a fetch handler — the server factory in
src/server.ts is transport-agnostic and stays unchanged.
Configuration
| Env var | Purpose | Default |
|---|---|---|
CODASMS_API_KEY |
Your coda_live_ reseller key (stdio mode). |
— (required) |
CODASMS_API_BASE |
API base URL. | https://api.codasms.com |
CODASMS_MAX_PRICE_CENTS |
Global cap for get_number, in US cents. |
unset |
PORT |
HTTP mode port. | 8080 |
Notes
- Rate limit: the API allows 60 requests/min per key;
wait_for_otppolls no faster than once every 5s by default to stay well under it. - Money safety:
get_numberspends real balance the moment it's called. If no code arrives, the order auto-refunds after 20 minutes, or callreleaseto refund immediately. - Codes: services and countries use the API's codes (e.g.
tg,wa, country6) — uselist_services/list_countriesto discover them.
Develop
npm install
npm run build # tsc -> dist/
npm start # stdio
License
MIT — see LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.