Keydris

Keydris lets AI agents spend money and change live systems while proving that every action was authorized - and lets you revoke that authority the instant it should end. It never holds your money or your private keys.

Why Keydris#

Autonomous agents increasingly take real actions: they move funds, call privileged APIs, and mutate production systems. The hard question is not can an agent act, but was this specific action authorized, and can I stop it right now? Keydris answers both.

• Prove authority. Every action an agent takes carries a signed token that says exactly what it may do and for how long.
• Enforce at the edge. The receiving system checks the token before the agent connects or acts - authorization is verified at the counterparty's front door, not on trust.
• Revoke at machine speed. When authority should end, it is pulled instantly across every boundary. A revoked token fails verification everywhere at once.
• Stay auditable. Every grant is written to an append-only, hash-chained log you can review later for compliance or dispute resolution.

The KIT (Keydris Identity Token)#

The KIT is a signed token issued to an agent. It declares what the agent may do and for how long, with minimum privilege as the default - an agent gets exactly the authority it needs for the task in front of it, and nothing more.

Status#

Keydris is currently in private beta (invite-only, limited early-adopter slots).

This is a demo documentation preview generated from the Keydris website to show how the product could look on Docsbook. It is not the official Keydris documentation.