> ## Documentation Index
> Fetch the complete documentation index at: https://docs.wiseguard.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Changelog

> What's new in the WiseGuard MCP connector.

<Update label="AI agent credits" description="October 2026">
  Tools that change something or send email now cost [AI agent credits](/guides/credits). Looking things up stays free.

  * Paid plans get a monthly allowance. Trials don't include credits.
  * Credits refill on the 1st of each month (UTC), and unused ones roll over up to 2× your monthly allowance.
  * You're only charged when a call succeeds.
  * `wiseguard_whoami` now reports how many credits you have left.
</Update>

<Update label="Mail tools" description="October 2026">
  Your AI agent can now read and send mail for your aliases, with two new scopes, `mail:read` and `mail:send`. If you connected before, [connect again](/guides/sign-in-and-permissions#what-you-approve) to approve them.

  * List and read the emails your aliases receive. Email content is marked as [untrusted](/guides/security-and-privacy#email-content-is-untrusted).
  * Reply from the alias an email was sent to, or send a new email from an active alias.
  * The same email is never sent twice, and agents have their own [send limits](/guides/limits-and-behaviour#sending-mail).
</Update>

<Update label="Alias tools" description="October 2026">
  Your AI agent can now manage your WiseGuard email aliases. Connect from [Claude or Claude Code](/quickstart).

  * Check your account and plan with `wiseguard_whoami`.
  * List, search and look up aliases.
  * Create aliases, with one alias per site.
  * Rename, pause and resume aliases.
  * Delete aliases, and restore them within 7 days.
</Update>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.