Glossary · AI governance
What is Open Knowledge Format?
Also known as: OKF, Open Knowledge Format standard
Open Knowledge Format, or OKF, is an open, vendor-neutral standard for packaging information as graph-linked Markdown that AI agents can read natively. Rather than locking knowledge inside proprietary systems, OKF structures it so that tools like AI assistants can traverse and reason over it directly.
Key takeaways
- OKF is an open, vendor-neutral standard for AI-readable knowledge.
- It packages information as graph-linked Markdown rather than proprietary formats.
- AI agents can read and traverse OKF content natively.
- CATAAM serializes your compliance program into OKF for tools like Claude.
What problem does OKF solve?
Most organizational knowledge lives in formats built for humans and proprietary tools: PDFs, spreadsheets, ticketing systems, and bespoke databases. AI agents struggle with this. They have to scrape, parse, and guess at relationships that were never made explicit, losing context and accuracy along the way.
Open Knowledge Format addresses this by packaging information in a way machines can consume directly. It expresses knowledge as Markdown, a simple and ubiquitous text format, and links those documents together into a graph so that relationships between concepts are explicit rather than implied. The result is knowledge that an AI agent can read and navigate natively, without brittle extraction steps.
Why graph-linked Markdown?
The two design choices work together. Markdown is open, human-readable, and trivially parseable, which makes the content transparent and portable rather than locked in a vendor format. Graph linking adds structure, connecting documents so an agent can follow relationships, for example from a policy to the controls that implement it to the evidence that proves it.
Because it is open and vendor neutral, OKF avoids the lock-in of proprietary knowledge stores. Anyone can read it, any compliant tool can consume it, and it travels with you. This makes it well suited to feeding context to AI assistants and agents that increasingly act on an organization's behalf.
How does CATAAM use OKF?
CATAAM serializes your compliance program into OKF, turning frameworks, controls, policies, tests, and evidence into graph-linked Markdown that AI tools can read directly. This means an assistant such as Claude can reason over your live compliance and AI governance program rather than working from stale, copied snapshots.
The practical payoff is an AI-native compliance program. Instead of a human exporting a report for an agent to digest, the program itself becomes the interface. Explore how this works on the OKF page.
Frequently asked questions
- How is OKF different from exporting a PDF or report for AI?
- A PDF or report is a flattened snapshot that an AI must parse and interpret, losing structure and context. OKF preserves relationships as a navigable graph of Markdown, so an agent reads the knowledge natively and can follow links between connected concepts.
- Why does AI-readable knowledge matter for compliance?
- As AI agents take on compliance and security tasks, they need accurate, current, structured context to reason correctly. Serializing a compliance program into OKF lets agents work from the live program directly, rather than from outdated copies, improving accuracy and reducing manual effort.
- Does using OKF lock me into one vendor?
- No. OKF is open and vendor neutral by design. Its use of plain Markdown and explicit graph links keeps the knowledge portable and readable by any compliant tool, which is the opposite of proprietary lock-in.