AI
OKF: Redefining Knowledge Bases for AI Agents
In June 2026, Google introduced the Open Knowledge Format (OKF), an open specification for structuring and exchanging knowledge bases used by AI agents.
Key takeaways
- OKF is an open specification for how AI agents organise and exchange knowledge.
- OKF uses Markdown files, lightweight YAML metadata, and links between concepts instead of embeddings or vector databases.
- OKF preserves document structure and relationships, enabling AI agents to navigate via direct links.
- OKF is version-controlled in Git, reviewable via pull requests, and searchable with standard tools.
- OKF is minimal and flexible, allowing consistent organization across domains like healthcare, software engineering, and manufacturing.
In June 2026, Google introduced the Open Knowledge Format (OKF), an open specification for structuring and exchanging knowledge bases used by AI agents. OKF uses plain Markdown files, lightweight YAML metadata, and explicit links between concepts instead of embeddings or vector databases. The format preserves document structure and relationships, enabling AI agents to navigate knowledge bases via direct links rather than similarity-based retrieval. OKF is designed to be version-controlled in Git, reviewed via pull requests, and searchable with standard tools. The specification is minimal, allowing flexible organization while maintaining consistency across domains like healthcare, software engineering, and manufacturing.
How it unfolded
- Google introduced OKF
Want this on autopilot?
We file the AI news, make the video, and post it for you.
Common questions
- marketing.blog.post.faq.what_happened
- OKF is an open specification for how AI agents organise and exchange knowledge.
- marketing.blog.post.faq.source
- marketing.blog.post.faq.source_answer
