What the Kanonik skills are

The signed skill library is what turns your own AI into a compliance worker. It loads over the connection, carries the methods, and needs nothing installed.

Kanonik does not give you an AI. You bring the one you already use, and Kanonik gives it a set of skills: the compliance methods a frontier model does not have on its own. The skills are the reason a general model becomes good at your ISMS work rather than merely fluent about it.

What a skill is

A skill is a package of instructions and reference material that teaches your AI how to do one kind of compliance work properly: how to draft a policy that maps to the framework, how to assess a control against real evidence, how to write an audit narrative an auditor will accept, and how to prepare a program for audit readiness. Loading a skill is what lets your assistant follow the method instead of improvising one.

The skills you load cover the core of the work:

• Policy drafting - producing policies that map to the requirements you are working toward.
• Control assessment - judging whether a control is designed and operating against the evidence on file.
• Audit narrative - writing the explanation behind an assessment in the form an auditor reads.
• Audit readiness - checking a program against what a specific framework's auditor will ask for.

They load on their own

There is nothing to install and nothing to copy into your AI client. Once you connect your AI to your workspace, the skills load over that connection automatically. From then on your assistant already knows the method, so you can ask for compliance work in plain language and get a first draft grounded in your real environment.

Private, signed, and yours to trust

The skill library is private and subscription-gated, not a public download, and every skill bundle is cryptographically signed. The version that was active when your AI produced a change is recorded in your tamper-evident audit log, so you can always show which method produced a given piece of work. The skills are instructions and reference material only: they carry no code and take no action on their own. Your AI does the reasoning; the skill tells it how to do that reasoning well; the Verifier and your approval decide what lands.

More help

Browse every article in the Help center, where you can also ask the Kanonik assistant directly. For anything else, email [email protected] and a person who works on the product answers.