In June 2026, Google launched the Open Information Format (OKF), an open specification for the way AI brokers organise and alternate data. An OKF bundle is simply Markdown information, light-weight YAML metadata, and hyperlinks between ideas, but it challenges the belief that each AI utility wants embeddings and vector databases.
As a result of the data base is obvious textual content, it may be version-controlled in Git and navigated by following hyperlinks reasonably than retrieving disconnected chunks. On this article, we’ll discover how OKF works and when it beats a standard retrieval pipeline.
Why Conventional RAG Has Limitations
Over the previous few years, Retrieval-Augmented Era (RAG) has develop into the usual strategy for offering exterior data to Giant Language Fashions. As an alternative of relying solely on the mannequin’s coaching knowledge, RAG retrieves related info from exterior paperwork throughout inference. A typical pipeline appears to be like one thing like this:
This strategy works remarkably effectively for looking tens of millions of paperwork. By evaluating the semantic which means of embeddings as an alternative of tangible key phrases, RAG permits an AI system to reply questions utilizing info that was by no means a part of the mannequin’s authentic coaching knowledge.
Nonetheless, there is a crucial trade-off. Earlier than a doc may be listed, it should first be divided into smaller chunks. Whereas chunking improves retrieval effectivity, it additionally breaks aside the unique construction of the doc. Relationships that had been naturally linked inside a single doc develop into distributed throughout a number of unbiased chunks.
Contemplate the next hospital protocol.
# Affected person Admission Coverage
Sufferers arriving via the Emergency Division should full an preliminary triage earlier than admission.
## Admission Necessities
– Legitimate affected person identification- Preliminary medical evaluation completed- Emergency instances obtain fast precedence
## Recording and Mattress Allocation
Affected person info is recorded within the Digital Well being File (EHR) system earlier than a mattress is assigned.
Mattress allocation follows the Mattress Occupancy tips maintained by the Operations group.
A typical RAG pipeline might cut up this doc into a number of smaller chunks earlier than indexing.
1st Chunk
Sufferers arriving via the Emergency Division should full an preliminary triage.
Admission Necessities:
– Legitimate affected person identification- Preliminary medical assessment- Emergency instances obtain fast precedence
2nd Chunk
Affected person info is recorded within the Digital Well being File (EHR) system.
third Chunk
Mattress allocation follows the Mattress Occupancy tips maintained by the Operations group.
Visually, the method appears to be like like this:
When a clinician asks,
“What’s the affected person admission course of?”
The vector database retrieves the chunks that appear most related, however the logical relationships between the admission coverage, emergency triage, the EHR system, and mattress allocation are misplaced. The mannequin has to reconstruct them on each question. This isn’t a flaw in RAG. It stays probably the greatest strategies for looking giant, unstructured collections like PDFs, analysis papers, help tickets, and historic information.
Curated organisational data is totally different. Insurance policies, procedures, APIs, and runbooks aren’t simply textual content, they’re interconnected ideas. Rebuilding these hyperlinks from fragmented chunks on each question provides pointless complexity, and that’s precisely the issue OKF was designed to unravel.
What’s the Open Information Format (OKF)?
The concepts behind OKF didn’t originate with Google. Earlier in 2026, Andrej Karpathy launched the idea of an LLM Wiki: as an alternative of repeatedly retrieving uncooked paperwork, an AI agent maintains a curated data base it might probably repeatedly learn, replace, and enhance. His analogy caught on shortly within the AI group:
Obsidian is the IDE. The LLM is the programmer. The wiki is the codebase.
The concept is easy. People present supply materials like documentation, insurance policies, schemas, and runbooks, and the agent organises it right into a structured wiki by writing summaries, connecting associated ideas, and sustaining hyperlinks. These relationships develop into a part of the data base as an alternative of being rediscovered on each question.
Google turned this group concept into an open specification. Relatively than delivery one other framework or SDK, it targeted on standardising the data itself. The result’s OKF, a light-weight format that shops data as unusual Markdown information with minimal metadata and express hyperlinks between ideas.
An OKF bundle is only a listing of Markdown paperwork, every representing one idea equivalent to a coverage, API, division, runbook, database desk, or metric, linked via commonplace Markdown hyperlinks. In contrast to a vector database that infers relationships via embedding similarity, OKF preserves them explicitly, so an agent follows hyperlinks reasonably than guessing.
As a result of all the things is obvious textual content, it matches current developer workflows: version-controlled in Git, reviewed by way of pull requests, and searchable with commonplace instruments. Subsequent, we’ll construct a bundle from scratch to see how it’s organised.
Construction of an OKF Bundle
Now that we’ve understood the motivation behind OKF, let’s take a look at how an OKF bundle is definitely organised. At its core, an OKF bundle is just a listing of Markdown information. Every Markdown file represents one idea, equivalent to a hospital coverage, division, process, system, or operational metric. Each idea accommodates light-weight metadata adopted by structured Markdown content material. Associated ideas are linked utilizing commonplace Markdown hyperlinks, permitting each people and AI brokers to navigate the data base naturally.
The specification itself is deliberately minimal. It defines only some conventions and avoids imposing a inflexible listing construction. This offers organisations the flexibleness to organise data in a means that most closely fits their area whereas nonetheless producing bundles that may be understood by any OKF-compatible agent.
A typical OKF bundle accommodates the next parts.
Part
Objective
index.md
Serves as the first entry level into the data base. It supplies an summary of the out there ideas and helps brokers navigate the bundle.
CHANGELOG.md (Elective)
Information modifications made to the data base over time, making updates clear and traceable.
Idea Recordsdata (.md)
Every Markdown file represents a single idea equivalent to a coverage, process, API, division, metric, or system.
YAML Entrance Matter
Shops metadata together with the idea sort, title, description, tags, possession, and final up to date timestamp.
Markdown Hyperlinks
Explicitly join associated ideas, reworking the data base right into a navigable graph as an alternative of remoted paperwork.
A Typical OKF Folder Construction
Though the OKF specification doesn’t mandate a specific listing structure, following a constant folder hierarchy makes the data base considerably simpler to keep up and navigate. The identical organisational ideas apply whatever the area.
The next examples display how totally different organisations can construction their data whereas following the identical OKF conventions.
1: Hospital Information Base
2: Software program Engineering Information Base
3: Manufacturing Information Base
Though these examples belong to fully totally different industries, the underlying organisation stays remarkably related. Each bundle begins with an index.md file that serves because the entry level, an non-compulsory CHANGELOG.md for monitoring revisions, and a set of directories that group associated ideas collectively.
This consistency is considered one of OKF’s greatest strengths. As soon as an AI agent understands how one OKF bundle is organised, it might probably navigate one other bundle constructed utilizing the identical conventions with little or no further adaptation.
Constructing an OKF Bundle
Now that we’ve explored the general construction of an OKF bundle, let’s construct one from scratch.
For the rest of this text, we’ll use a fictional hospital known as CityCare Hospital. Think about we’re constructing an AI assistant that helps docs, nurses, and hospital directors reply operational questions. The assistant ought to perceive admission insurance policies, emergency procedures, hospital departments, inner methods, and operational metrics. As an alternative of storing this info inside a vector database, we’ll organise it as an OKF bundle.
We’ll start by creating the basis listing.
The index.md file acts because the entry level for each people and AI brokers.
# CityCare Hospital Information Base
## Insurance policies
– [Patient Admission Policy](insurance policies/patient-admission.md)- [Discharge Policy](insurance policies/discharge-policy.md)
## Procedures
– [Emergency Triage](procedures/emergency-triage.md)- [Blood Transfusion](procedures/blood-transfusion.md)
## Techniques
– [Electronic Health Record](methods/ehr-system.md)
## Metrics
– [Bed Occupancy](metrics/bed-occupancy.md)
## Departments
– [Emergency Department](departments/emergency.md)
Relatively than looking the whole repository, an AI agent can first learn the index to grasp what ideas exist earlier than navigating to the related information. This straightforward design retains the bundle organised whereas decreasing pointless context throughout retrieval.
Within the subsequent part, we’ll create particular person idea information and study how YAML metadata, Markdown content material, and hyperlinks work collectively to make the data base comprehensible for each people and AI brokers.
Creating an OKF Idea File
The constructing blocks of an OKF bundle are idea information. Every idea represents precisely one piece of information, equivalent to a coverage, process, division, system, metric, or API. Retaining ideas targeted makes them simpler to keep up whereas permitting AI brokers to retrieve solely the data they want.
Each idea file consists of two elements:
YAML Entrance Matter, which shops metadata concerning the idea.
Markdown Content material, which accommodates the precise data together with hyperlinks to associated ideas.
Let’s create an idea file for the hospital’s affected person admission coverage.
—type: policytitle: Affected person Admission Policydescription: Pointers for admitting sufferers into CityCare Hospitaltags:- admissions- patient-careupdated: 2026-06-15—
# Affected person Admission Coverage
Sufferers arriving via the Emergency Division should full an preliminary triage earlier than admission.
## Admission Necessities
– Legitimate affected person identification- Preliminary medical evaluation completed- Emergency instances obtain fast precedence
## Associated Ideas
– [Emergency Triage](../procedures/emergency-triage.md)- [Electronic Health Record](../methods/ehr-system.md)
Discover that the file accommodates way more than plain textual content. The YAML part describes what sort of data this file represents, whereas the Markdown physique explains the idea intimately. Most significantly, the idea hyperlinks to different associated ideas contained in the data base. These hyperlinks rework remoted paperwork into an interconnected data graph that an AI agent can navigate.
Though OKF solely requires the sort discipline, including further metadata makes the bundle simpler to organise and preserve.
Area
Description
sort
Identifies the kind of idea, equivalent to coverage, process, system, or metric. That is the one required discipline within the present specification.
title
Human-readable title of the idea.
description
Abstract describing the idea.
tags
Key phrases that assist organise associated ideas.
up to date
Signifies when the idea was final modified.
Because the metadata is saved in YAML, each people and AI brokers can shortly perceive what a doc represents earlier than studying its full content material.
One of many greatest variations between OKF and conventional doc storage is that ideas are explicitly linked utilizing Markdown hyperlinks.
For instance, the admission coverage references the emergency triage process and the Digital Well being File (EHR) system.
These relationships are deliberately created by the creator. The agent doesn’t should infer them via semantic similarity as a result of they exist already contained in the data base.
One other Instance: Hospital Metric
Idea information usually are not restricted to insurance policies. The identical construction can describe operational metrics, inner methods, departments, APIs, or runbooks.
Beneath is an idea describing the hospital’s Mattress Occupancy Price.
—type: metrictitle: Mattress Occupancy Ratedescription: Proportion of inpatient beds presently occupiedtags:- operations- hospitalupdated: 2026-06-15—
# Mattress Occupancy Price
The Mattress Occupancy Price measures the share of inpatient beds presently occupied.
## FormulaOccupied Beds / Complete Out there Beds × 100
## Knowledge Supply
Hospital Data System
## Proprietor
Operations Division
## Associated Ideas
– [Emergency Department](../departments/emergency.md)- [Patient Admission Policy](../insurance policies/patient-admission.md)
As a result of each idea follows a constant construction, an AI agent can shortly perceive what the metric represents, how it’s calculated, the place the info comes from, and which different ideas are associated to it.
How AI Brokers Traverse an OKF Bundle
As soon as the data base is organised into interconnected idea information, retrieval turns into a lot easier than conventional doc search.
As an alternative of looking 1000’s of doc chunks, an AI agent follows a structured navigation course of.
Learn the index.md file to grasp the general data base.
Determine essentially the most related idea primarily based on the consumer’s query.
Open that idea file.
Observe hyperlinks to associated ideas each time further context is required.
Generate the ultimate response utilizing solely the related ideas.
The traversal course of may be visualised as follows.
In contrast to a RAG pipeline, the agent doesn’t start by looking an embedding index. It begins from a curated entry level and progressively explores solely the ideas which might be related to the present activity.
This strategy preserves the relationships between ideas whereas protecting the quantity of context despatched to the language mannequin comparatively small.
Why This Works Properly for AI Brokers
The largest benefit of OKF is that it permits builders to organise data in the identical means people naturally give it some thought. A physician studying the hospital’s documentation doesn’t randomly leap between unrelated paragraphs. They start with a coverage, observe references to procedures, seek the advice of the related methods, after which arrive on the info they want. OKF allows AI brokers to observe this identical workflow.
As an alternative of reconstructing relationships from fragmented doc chunks each time a query is requested, the agent navigates an express data graph the place these relationships have already been outlined. This makes the retrieval course of extra deterministic, simpler to audit, and considerably easier to keep up.
The place RAG Nonetheless Excels
At this level, OKF seems like a really perfect resolution for organising data. It preserves relationships between ideas, retains all the things version-controlled, and permits AI brokers to navigate curated documentation with out counting on semantic search.
Nonetheless, OKF has an vital limitation. Somebody has to curate each idea.
Each coverage, process, system, metric, and division have to be written, reviewed, and maintained. This works effectively for authoritative organisational data, but it surely turns into impractical when the data base grows to tens of millions of paperwork.
Contemplate a hospital that has accrued years of operational knowledge.
Hundreds of thousands of Digital Well being File (EHR) entries
Medical analysis papers
Scientific notes
Affected person suggestions
Incident stories
Inside emails
Assembly transcripts
Gear upkeep logs
Organising each considered one of these paperwork into fastidiously curated OKF idea information would require an infinite quantity of handbook effort. Even when AI brokers assisted with the curation course of, a lot of this info modifications repeatedly and is best fitted to semantic search.
That is the place Retrieval-Augmented Era (RAG) continues to be the popular resolution.
RAG Resolution
Relatively than requiring paperwork to be manually organised, RAG indexes giant collections of unstructured knowledge utilizing embeddings. When a query is requested, the system retrieves essentially the most semantically related paperwork and supplies them as context to the language mannequin.
For instance, think about the next questions:
Has anybody encountered this MRI scanner error earlier than?
Discover earlier incident stories involving delayed laboratory outcomes.
Summarise discussions concerning the new EHR rollout.
Seek for all assembly notes mentioning affected person switch delays.
These questions can’t be answered from a small curated data base. As an alternative, they require looking via 1000’s and even tens of millions of paperwork the place the reply may exist anyplace. That is precisely the kind of drawback RAG was designed to unravel.
The strengths of every strategy develop into a lot clearer when considered aspect by aspect.
Function
OKF
RAG
Finest for
Curated organisational data
Giant collections of unstructured paperwork
Information Supply
Markdown idea information
Uncooked paperwork
Retrieval
Deterministic navigation
Semantic similarity search
Infrastructure
File system + Git
Embeddings + Vector Database
Model Management
Native Git help
Requires re-indexing after updates
Relationships
Specific hyperlinks between ideas
Inferred from retrieved chunks
Scalability
Average
Wonderful
Explainability
Excessive
Average
Neither strategy is universally higher than the opposite. They merely remedy totally different issues. OKF supplies construction and precision. RAG supplies scale and suppleness. This naturally raises one other query.
Do we actually have to decide on one over the opposite?
Fortuitously, the reply is not any.
Hybrid Information Structure: Combining OKF and RAG
In observe, the simplest AI methods use each OKF and RAG collectively.
As an alternative of treating them as competing applied sciences, fashionable agent architectures use every one the place it performs finest.
A easy means to consider that is the 80/20 precept.
The 20% of organisational data that’s crucial, steady, and regularly referenced is saved as an OKF bundle.
The remaining 80% of huge, unstructured info stays inside a standard RAG pipeline.
This creates a layered data structure.
The router determines which data supply is most applicable for the incoming question.
Questions requiring authoritative and deterministic solutions are routed to the OKF bundle.
Examples embrace:
What’s the affected person admission coverage?
How is the Mattress Occupancy Price calculated?
Which system shops affected person medical information?
What’s the emergency blood transfusion process?
Every of those questions has a single authoritative reply maintained by the organisation.
Then again, exploratory questions are routed to the RAG pipeline.
For instance:
Has anybody encountered this MRI scanner error earlier than?
Discover earlier incident stories involving delayed laboratory outcomes.
Summarise discussions concerning the EHR migration challenge.
Search assembly notes discussing affected person discharge delays.
These questions require looking giant collections of historic paperwork reasonably than consulting curated data.
This hybrid structure permits every system to concentrate on its strengths.
OKF Strengths
RAG Strengths
Deterministic retrieval
Semantic retrieval
Curated and authoritative data
Large doc collections
Specific relationships between ideas
Finds info utilizing semantic similarity
Model-controlled with Git
Constantly indexes new paperwork
Straightforward to audit and preserve
Extremely scalable
Maybe the most important benefit of this structure is that the language mannequin doesn’t have to know the place the data comes from. The agent merely requests the data it wants.
The routing layer decides whether or not that data ought to come from the OKF bundle or the vector database. Frameworks equivalent to LangGraph, LangChain, or LlamaIndex make this routing easy by permitting builders to construct workflows that select the suitable retrieval technique primarily based on the consumer’s question. In consequence, AI brokers achieve the precision of curated data with out sacrificing the flexibility to look huge collections of unstructured info.
In different phrases, the long run will not be OKF versus RAG. It’s OKF plus RAG, working collectively as complementary layers in a single data structure.
Conclusion
The Open Information Format presents a easy, clear approach to organise data for AI brokers. By representing it as interconnected Markdown paperwork, OKF retains organisational data simple to grasp, preserve, and version-control in Git.
It fits curated info like insurance policies, runbooks, and API docs, with out changing RAG, which nonetheless excels at semantic search throughout giant, unstructured collections.
Used collectively, the 2 cowl way more floor than both alone. In the end, understanding the place every strategy matches is what lets builders construct brokers which might be correct, explainable, and production-ready.
Ceaselessly Requested Questions
A. It’s an open specification Google launched in June 2026 for the way AI brokers organise and alternate data. An OKF bundle is a listing of Markdown information with light-weight YAML metadata and express hyperlinks between ideas, so data lives as plain textual content alongside your code reasonably than as vectors in a database.
A. RAG splits paperwork into chunks, embeds them, and retrieves essentially the most semantically related items at question time. OKF shops data as interconnected idea information and lets an agent navigate by following author-defined hyperlinks. RAG infers relationships; OKF retains them express.
A. No. They remedy totally different issues. OKF is finest for curated, authoritative data like insurance policies, runbooks, and API docs. RAG is finest for looking giant, unstructured collections equivalent to EHR entries, incident stories, and assembly notes. The article recommends utilizing them collectively.
Knowledge Scientist @ Analytics Vidhya | CSE AI and ML @ VIT ChennaiPassionate about AI and machine studying, I am wanting to dive into roles as an AI/ML Engineer or Knowledge Scientist the place I could make an actual influence. With a knack for fast studying and a love for teamwork, I am excited to deliver revolutionary options and cutting-edge developments to the desk. My curiosity drives me to discover AI throughout numerous fields and take the initiative to delve into knowledge engineering, making certain I keep forward and ship impactful tasks.
Login to proceed studying and revel in expert-curated content material.
Preserve Studying for Free
