Redis leased credentials, verified connections, a grant builder, and Discord and Slack alerts

Three additions to leased credentials, two for webhooks, and a new way to export your audit log.

Redis

Issue short-lived Redis logins the same way you do for PostgreSQL, MySQL, and MariaDB. Point SikkerKey at your Redis instance, choose the access each login gets, and every machine receives its own credential that expires and revokes on its own.

Connection security

Every leased connection now has a clear security level. Encrypted keeps the connection private; Encrypted + verified also verifies your database's TLS certificate, confirming the server's identity. Databases with publicly trusted certificates verify automatically, and you can paste a certificate for self-signed or private certificate authorities.

Grant builder

A new Build template option generates the access each login receives from a preset, individual privileges, and a scope, then writes the correct statement for your provider, including the parts that are easy to miss by hand. You can also edit grants after a leased secret is created.

See the leased credentials documentation for the details.

Webhook setup

Adding a webhook is now a guided flow: choose an integration, point it at its destination, pick the events to send, and review before it goes live. Each webhook also has a settings panel for updating its events and checking delivery health.

Discord and Slack

Send the audit events you choose straight to a Discord or Slack channel, whether that's a sign-in, a secret change, a new machine, or a role update. Each one arrives as a message, color-coded by severity, carrying the details that apply to that event.

Audit log export

Export your audit log in the format that fits: CSV for a spreadsheet, JSON for your own tooling, or plain text to read or attach to a report. Choose a date range, apply the filters you want, and download every event that matches, however far back it reaches.

Filter by action, severity, or source IP, search the event details, and narrow by who acted, either a whole actor type or a specific machine, user, or AI identity. The JSON and text files record the range and filters inside them, so every export describes itself.