What onelinktoken is, how the REST API works, and the seven tools an AI agent gets over MCP.
onelinktoken issues, delivers, validates, and revokes short-lived, single-use credentials — magic links, one-time passwords, and bearer tokens — over a REST API, an MCP server, and webhook callbacks. We store a SHA-256 hash of each token value rather than the value itself, so a token can be validated but never read back out of our systems.
It is not an identity provider, a user directory, or a system of record for your users — it is the token layer that sits underneath your own auth: mint a link or code, hand it to a user or an agent, and revoke it the moment you are done with it.
Every call is authenticated with a bearer API key and namespaced under /api/v1
on onelinktoken.com. There are three core operations:
Create a magic link, OTP, or bearer token — pass token_type, a target, and an expiry.
Validate and spend a token once. A redeemed token cannot be reused.
Kill a token immediately, whether or not it has been used. Propagates in <100ms.
Status and listing operations are also available — see get_token_status and
list_tokens in the MCP tool catalog below for the same operations exposed as
MCP tools.
# POST /api/v1/tokens curl -X POST https://onelinktoken.com/api/v1/tokens \ -H "Authorization: Bearer tf_key" \ -H "Content-Type: application/json" \ -d '{ "token_type": "magic_link", "target_email": "user@hospital.org", "expires_in_minutes": 15, "max_uses": 1 }' # → { "id": "tok_abc", "magic_link_url": "https://onelinktoken.com/r/...", "status": "pending" }
Everything the REST API does is also exposed as MCP tools, so an AI agent can discover and call them directly instead of writing an API call. The connector advertises exactly these seven tools:
Mints a single-use sign-in link for an address, with an expiry you choose.
Mints a short numeric one-time password for an email or SMS verification step.
Mints an API access token that can be revoked the moment you are done with it.
Spends a token once and reports whether it was valid. A redeemed token cannot be reused.
Kills a token immediately, whether or not it has been used.
Reports the state of one token — pending, redeemed, expired, or revoked.
Lists your recent tokens and their states so you can audit or clean up.
Adding onelinktoken as a remote MCP connector to Claude or ChatGPT, the OAuth consent flow, free-tier limits, and — importantly — how to handle the fact that any token minted in a chat ends up in that chat's transcript, are all covered in full on the connect page.
Endpoint details, step-by-step setup for Claude and ChatGPT, what the OAuth consent screen shows you, and the chat-transcript privacy warning you should read before minting anything sensitive.
Read the connect guide →There is no hosted signup yet. onelinktoken-server is open and self-hostable: clone the repo, configure it with your own database, cache, and mail provider, and run the API and MCP server on your own infrastructure.
Setup instructions, configuration options, and deployment notes live in that repo's own README.
onelinktoken-server on GitHub →For anything not covered here — integration help, billing, or a security report — see Support.