Best specification OKF Bundles for AI Agents
Specification OKF bundles provide structured knowledge for AI agents working in this domain. These bundles are designed to be loaded directly into agent context, giving your AI assistant the domain expertise it needs without manual setup or complex infrastructure. Browse the top bundles below to find the right one for your workflow.
Top specification OKF Bundles
MIF
Open standard for portable AI knowledge interchange — JSON-LD + Markdown dual format with W3C PROV provenance, bi-temporal tracking, entity relationships, and OKF support.
Google Knowledge Catalog (OKF Reference)
OKF ✓The official Google Cloud Platform reference implementation of OKF. Includes the format specification, a reference agent that produces OKF bundles from BigQuery datasets, and three example bundles (GA4 e-commerce, Stack Overflow, Bitcoin blockchain).
ssss
SSSS — a database-free, Markdown-first schema and mutation contract for AI-agent state. Turn a running business into a single tradeable .ucw file.
Related Topics
Frequently Asked Questions
What is a specification OKF bundle?
A specification OKF bundle is a portable knowledge package in the Open Knowledge Format that AI agents can load and use. These bundles contain structured markdown documents with YAML frontmatter, organized around specification use cases. They work with Claude Code, Cursor, OpenCode, and any OKF-compatible agent — no SDKs, databases, or API keys required.
Which specification bundle is best for Claude Code?
The best specification bundle for Claude Code depends on your specific needs. The bundles above are ranked by GitHub stars — a good proxy for community adoption. Look for bundles with OKF conformance status and recent updates. Claude Code natively supports OKF bundles, so any conformant bundle will work out of the box.
Are these bundles free?
Yes, all OKF bundles on BundleDex are free and open source. They are hosted on public GitHub repositories and can be cloned, forked, and used without any licensing fees or API keys. Just git clone the repository and point your AI agent at the directory.
Don't see what you're looking for? Browse all bundles or submit your own.