Intelligence Briefing · Viska Funds

The Open Knowledge Format

Google Cloud's new open specification for storing an organisation's knowledge as a folder of linked Markdown files — what it is, what it is not, and why the idea matters regardless of whether this particular standard succeeds.

Prepared for
Viska — fund decision-makers
Prepared by
Hermes · intelligence pipeline
Date
24 June 2026
Research tier
T-DEEP · primary-source verified

Scope: Neutral landscape briefing. No Viska-specific recommendation — this document explains OKF so the fund can form its own view.

Section 01

Executive summary

On 12 June 2026, Google Cloud published the Open Knowledge Format (OKF) v0.1 — an open specification for storing an organisation's knowledge as a folder of plain Markdown files that link to each other, rather than inside a specialised "vector database." It formalises a pattern popularised two months earlier by Andrej Karpathy (co-founder of OpenAI, former head of AI at Tesla), who called it the "LLM Wiki."

The core idea is a genuine shift — and it is the part worth your attention regardless of whether this particular spec succeeds:

The dominant way to give an AI system memory has been RAG — chop documents into fragments, store them in a vector database, and retrieve the closest fragments to each question. Its weakness: the system re-derives the same connections on every query; nothing accumulates. The "wiki" approach instead has an AI build and maintain a living, cross-linked knowledge base once, so the synthesis is paid for up front and simply read thereafter.

Bottom line for decision-makers

The idea is sound & spreading

Essentially "keep institutional knowledge as version-controlled Markdown that an AI maintains." Low-cost, low-lock-in, inspectable.

The standard is early & minimal

A v0.1 draft. It standardises the container (file/folder conventions) but deliberately not the meaning of the content — its most-cited weakness.

Not magic, not a DB replacement at scale

Even Karpathy reintroduces conventional search (incl. vector search) once a base grows past a few hundred documents. "A folder beats the vector database" holds at moderate scale, not universally.

Genuinely low-risk to evaluate

Apache-2.0 licence, no vendor account or SDK required to read or write, lives in Git. Trying it costs little and locks you into nothing.

Section 02

What OKF actually is, in plain terms

An OKF "bundle" is just a directory of Markdown files. Each file is one concept — a table, a metric, a playbook, a process, an entity, anything. The file's location is its name. Files link to one another with ordinary Markdown links, and those links form a knowledge graph.

The whole specification fits on one page. In Google's own words: "If you can cat a file, you can read OKF; if you can git clone a repo, you can ship it."

There is exactly one hard rule: every file must declare a type: in a small YAML header (e.g. type: Metric). Everything else — title, description, source link, tags, timestamp — is recommended but optional. Two filenames are reserved:

index.md

A table of contents for a folder. An AI reads this first to see what exists before opening individual files ("progressive disclosure").

log.md

A dated changelog of what was added or changed.

A deliberately permissive rule governs anything reading a bundle: it must tolerate unknown fields, unknown type values, broken links, and missing index files, and must never reject a bundle for them. Google states this is intentional — bundles will be "partially generated by agents," so readers are told to forgive mess rather than demand perfection.

Section 03

Where it came from

APRIL 2026

Andrej Karpathy's "LLM Wiki"

Karpathy published a short "idea file" describing the pattern. His framing, verbatim: "Obsidian is the IDE; the LLM is the programmer; the wiki is the codebase." The key insight is that the human does not write the wiki — the AI does: it reads each new source, updates entity pages, revises summaries, flags where new data contradicts old, and maintains the cross-references. The human curates sources and asks good questions; the AI does the bookkeeping that humans always abandon.

Crucially, Karpathy's pattern has three layers: immutable raw sources; the AI-written wiki; and a schema file (a CLAUDE.md/AGENTS.md-style document) that tells the AI how to maintain the wiki — plus three standing operations: Ingest, Query, and Lint (a periodic health-check for contradictions, stale claims, and orphaned pages).

12 JUNE 2026

Google formalises it as OKF

Google Cloud's announcement explicitly says OKF "formalizes the LLM-wiki pattern" and credits Karpathy directly. Google's contribution is standardisation — pinning down the minimum file/folder conventions so wikis written by one team or tool can be read by another.

Important nuance

Google standardised the container (the file format) but left out Karpathy's maintenance machinery — the schema layer and the Ingest/Query/Lint operations that actually keep a wiki alive. The OKF spec is a static format definition; it contains no mechanism for keeping bundles current. (See Section 06, limitation 1.)

Section 04

What Google actually shipped

OKF is published openly on GitHub (GoogleCloudPlatform/knowledge-catalog, Apache-2.0). Alongside the one-page spec, Google shipped three reference implementations, explicitly labelled "proofs of concept":

01

Enrichment agent

Walks a BigQuery dataset, drafts an OKF document for every table/view, then runs a second AI pass to add citations, schemas, and join paths.

02

Static HTML visualiser

Turns any bundle into an interactive graph view in a single self-contained file — no backend, "no data leaves the page."

03

Three sample bundles

Built from BigQuery public datasets: Google Analytics e-commerce, Stack Overflow, and Bitcoin.

Authorship is telling about strategy: the announcement is by-lined by Google's Data Cloud / BigQuery engineering leads, not its AI research lab, and every example is BigQuery-flavoured. Google has also wired its commercial Knowledge Catalog product to ingest OKF. The format is open and genuinely usable by anyone; the reference tooling and samples point back toward Google's data platform. That is a normal "open standard, house tooling" play — worth understanding, not alarming.

⚠ One claim we could not verify

A widely-shared video about OKF asserts Google "just renamed" one of its knowledge products to capitalise on this launch. We found no evidence of any such rename. The product Google actually connected to OKF is its existing Knowledge Catalog. We flag this so the claim is not repeated as fact.

Section 05

The case for it — why people are excited

  1. Synthesis is paid once, not per query.

    A vector-RAG system re-assembles the answer to "how do we compute weekly active users?" from scattered fragments every single time. A maintained wiki has already written that page, already flagged the contradictions, already linked the related concepts. You read a finished answer instead of re-deriving it.

  2. It scales by navigation, not brute force.

    An AI reads the small index.md table-of-contents first and opens only the one or two files it needs — it never has to ingest the entire library at once.

  3. It is just text in Git.

    No database, server, or API key is needed to read it. You can diff it, review changes in a pull request, hand the whole thing to an offline model, and read it in any text editor. Knowledge becomes auditable and portable in a way a vector database never is.

  4. Low lock-in.

    Open spec, Apache-2.0, no SDK. The data outlives any one tool or vendor.

Section 06

The limitations, graded honestly

We assessed the three most-cited criticisms against the actual spec and independent commentary. Our confidence in each is stated.

Limitation 1 — There is no update mechanism.

Verified · high confidence

The spec has a timestamp field, but a field is not a process. OKF defines a container and says nothing about how bundles stay current — Google explicitly dropped the maintenance layer that Karpathy's original included. In practice this works beautifully when one owner tends one folder; on a shared team knowledge base nobody is assigned to maintain, it goes stale, and an AI then answers confidently from expired knowledge.

Mitigation exists (it is exactly Karpathy's schema + Ingest/Query/Lint discipline, or a scheduled maintenance job) — but you must supply it; the standard will not.

Limitation 2 — It standardises the box, not the meaning.

Verified · high · corroborated

The single required field, type, is free-form with no shared vocabulary. One team writes type: BigQuery Table, another writes table, another relational asset — all valid, all conformant, none interoperable in meaning. An independent analyst put it precisely: "The container is standardized; the meaning is left to each producer … a shared way to store context, not yet a shared way to make sense of it."

That same analysis caught an early inconsistency: Google's own reference parser expects four fields even though the spec requires only one — a sign the standard is still settling. For a single organisation that controls its own vocabulary, this matters far less; for cross-organisation exchange (OKF's stated ambition), it is the open problem.

Limitation 3 — AI "librarians" can be unreliable at scale.

Plausible · medium · partly inferred

The concern: language models maintaining thousands of files can botch formatting, mangle structure, and invent links to files that do not exist. We did not find a hard, quantified study proving this at scale, so we grade it as a reasonable risk rather than an established fact. It is, however, implicitly conceded by the spec's design — the "tolerate all mess" rule exists precisely because bundles are "partially generated by agents."

For a fund, the sharper version of this risk is factual: an AI that fabricates a number while summarising a financial document is far more dangerous than one that mis-formats a heading. Any serious deployment needs provenance discipline (OKF does provide an optional # Citations convention to support this) and validation — neither of which the format enforces.

A fourth, under-discussed caveat — scale.

Originator-stated

Karpathy himself scopes the pure index-first approach to "moderate scale (~100 sources, ~hundreds of pages)." Beyond that he recommends adding a real search engine — specifically qmd, which uses hybrid keyword + vector search with re-ranking. In other words, the originator reintroduces vector search once the corpus is large. The popular "a folder beats the vector database" headline is therefore misleading at high volume: at scale the mature pattern is folder plus search, not folder instead of search.

Section 07

Adoption & maturity

~5,000
GitHub stars
381
Forks
45
Open issues
31
Open pull requests

Status: v0.1 — explicitly a "starting point, not a finished standard." Google invites breaking changes in future versions.

Traction: the figures above are ~12 days after launch — strong early interest and real engagement, and they contradict the dismissive "a standard with one user" framing. Caveat: stars measure curiosity, not production adoption; it is far too early to call OKF an established standard, and it could still stall (Google has retired promising projects before).

Licence: Apache-2.0 (permissive, commercially safe). The repo carries the standard "not an official Google product" disclaimer — it is a reference/sample repository, not a supported product.

Backing: Google Cloud Data Cloud / BigQuery engineering, published in the open with explicit invitation for outside implementations.

Section 08

How it relates to the alternatives

ApproachWhat it isStrengthWeakness
Vector RAGDocuments chunked, embedded, retrieved by similarity per queryFine-grained recall over huge, unstructured corpora; mature toolingRe-derives connections every query; nothing accumulates; opaque
OKF / LLM-WikiAI-maintained folder of linked Markdown conceptsSynthesis persists; auditable; portable; no infra to readNo built-in maintenance; meaning not standardised; weaker at large-scale fine recall
Knowledge graphTyped entities + typed relationships in a graph storeStrongest at precise relationship / semantic queriesHeavyweight; needs schema design + infrastructure
MCP (Model Context Protocol)A protocol for connecting AI to live tools / dataStandard plumbing for real-time accessDifferent layer entirely — not a memory format

"MCP is the pipe, OKF is the cargo."

Commentators' framing, not Google's official wording. MCP moves data to a model at run time; OKF is a way to store curated knowledge — they are complementary, not competing. Likewise, OKF and vector RAG are not strictly either/or: the emerging mature pattern (and Karpathy's own) is a maintained wiki for synthesised knowledge, with a search index (often hybrid keyword + vector) layered on for scale.

Section 09 · Bottom line

Google blessing an already-spreading good idea

OKF is best understood as Google blessing and standardising an already-spreading good idea: keep your institutional knowledge as AI-maintained, version-controlled Markdown. The idea is durable and worth internalising — it directly attacks the real weakness of today's RAG systems (no memory, constant re-derivation). The specific v0.1 standard is minimal, early, and solves the easy half of the problem (the container) while explicitly leaving the hard halves — keeping knowledge current, and agreeing on what words mean — to whoever adopts it.

For any knowledge-heavy organisation, the responsible posture is: take the pattern seriously; treat the v0.1 standard as promising-but-unproven; and assume you must supply the maintenance discipline, provenance controls, and (at scale) search infrastructure that the format itself does not provide.

Section 10 · Appendix

Source-coverage ledger

Every load-bearing claim, with how it was verified. This briefing is T-deep: each primary source below was read in full.

Tags — deep-read read the full primary source · secondary credible third party · inferred reasoned from evidence · unverified could not confirm.

20 verified claims + primary sources expand
#ClaimTagSource
1OKF published 12 June 2026, v0.1 draftdeep-readGoogle Cloud blog (first_published 2026-06-12)
2OKF = directory of Markdown + YAML frontmatter; bundle / concept / concept-ID modeldeep-readOKF SPEC.md §1–§4
3Exactly one required field: type; not centrally registered; consumers must tolerate unknown typesdeep-readSPEC.md §4.1, §9; blog
4Reserved files are index.md (TOC) + log.md (changelog)deep-readSPEC.md §3.1, §6, §7
5Permissive consumption: must not reject for unknown fields/types, broken links, missing indexdeep-readSPEC.md §9
6Provenance supported via optional # Citations conventiondeep-readSPEC.md §8
7Spec contains no maintenance / update mechanism (Limitation 1)deep-readSPEC.md (entire) — absence verified; blog confirms format-only scope
8Origin = Karpathy "LLM Wiki," ~April 2026; AI writes the wiki, not the humandeep-readKarpathy gist (raw)
9Exact quote "Obsidian is the IDE; the LLM is the programmer; the wiki is the codebase"deep-readKarpathy gist
10Karpathy's three layers (raw/wiki/schema) + Ingest/Query/Lint = the maintenance layer OKF droppeddeep-readKarpathy gist
11Karpathy scopes index-first nav to "~100 sources"; recommends qmd (hybrid BM25/vector) at scaledeep-readKarpathy gist
12Google blog "formalizes the LLM-wiki pattern," credits Karpathydeep-readGoogle Cloud blog
13Authors = Sam McVeety (Data Analytics TL) + Amir Hormati (BigQuery TL) → "BigQuery-team origin"deep-readGoogle Cloud blog byline
14Reference impls: BigQuery enrichment agent, static HTML graph visualiser, 3 BigQuery sample bundlesdeep-readGoogle Cloud blog
15Adoption ~12d post-launch: ~5,000 stars, 381 forks, 45 issues, 31 PRs, Apache-2.0secondaryGitHub repo page (fetched 2026-06-24)
16Container-not-meaning critique; Google reference parser wants 4 fields vs spec's 1secondaryMarc Bara, Medium (independent analysis)
17"Renamed knowledge product" claim from source videounverifiedNo corroborating source; blog points to existing Knowledge Catalog
18Reference enrichment agent "runs on Gemini"inferredBlog says "a second LLM pass," does not name Gemini
19"MCP is the pipe, OKF is the cargo" positioningsecondaryCommentator framing; not in Google's official text
20Limitation 3 (LLMs unreliable as librarians at scale)inferredNo hard study found; supported circumstantially by spec's "tolerate mess" rationale

Primary sources (read in full)

  • OKF specification — raw.githubusercontent.com/GoogleCloudPlatform/knowledge-catalog/main/okf/SPEC.md
  • Google Cloud announcement — cloud.google.com/blog/products/data-analytics/how-the-open-knowledge-format-can-improve-data-sharing
  • Karpathy "LLM Wiki" gist — gist.github.com/karpathy/442a6bf555914893e9891c11519de94f
  • Reference repo (adoption metrics) — github.com/GoogleCloudPlatform/knowledge-catalog
  • Independent critique — Marc Bara, "Google's New Format for Agent Context: A Standard, or Just a Folder?", Medium

Trigger source (secondary, not authoritative)

Devsplainers, "Google's OKF: Why a Folder Beats the Vector Database" (YouTube, 2026-06-22). This video prompted the research; two of its specific claims (the "renamed product"; the reserved-file naming) did not survive primary-source verification — see the audit below.

Verification appendix

Trigger-video claim audit

The research was prompted by a YouTube video. Per verify-before-assert discipline, the video was treated as an unverified secondary source and each of its specific claims independently audited — so the briefing inherits none of its mistakes.

Claim-by-claim audit (✅ correct · ⚠️ partial · ❌ wrong) collapse
Video claimVerdictEvidence
OKF published 12 June✅ CorrectBlog first_published 2026-06-12
Karpathy posted "LLM Wiki" in April; OpenAI co-founder / ex-Tesla AI✅ CorrectGist + public record
"Obsidian is the IDE, the LLM is the programmer, the wiki is the codebase"✅ Correct (exact)Karpathy gist, verbatim
Bundle = folder; one file = one concept; path = name; links = graph✅ CorrectSPEC.md §2–§5
One required field; readers must forgive unknown fields / broken links / unparseable files✅ CorrectSPEC.md §4.1, §9
Google kept the folder but dropped Karpathy's maintenance instructions✅ CorrectGist has schema + Ingest/Query/Lint; SPEC.md has none
Critique: no update process (field ≠ process); shared folders go stale✅ CorrectSPEC.md has no maintenance mechanism
Critique: standardises container, not meaning (free-form type)✅ CorrectSPEC.md §4.1; corroborated by Marc Bara
OKF came from the BigQuery team; samples on BigQuery✅ Essentially correctAuthors = Data Analytics + BigQuery TLs; all samples BigQuery datasets
Two special files (one TOC, one changelog)⚠️ Right idea, wrong namesActual reserved names are index.md + log.md; video implied bespoke two-letter names
Reference writer "runs on Gemini"⚠️ UnconfirmedBlog says "a second LLM pass," never names Gemini
Critique: LLMs are bad markdown librarians at scale⚠️ Plausible, not provenNo hard study found; only circumstantial support
"Almost nobody outside Google was using it / a standard with one user"⚠️ Overstated~5,000 stars / 381 forks / 31 PRs in ~12 days = strong early interest
Google "just renamed" a knowledge product for this launch❌ Likely wrongNo evidence of any rename; product wired to OKF is the existing Knowledge Catalog
"A folder beats the vector database"❌ Misleading at scaleKarpathy reintroduces hybrid keyword+vector (qmd) beyond ~hundreds of pages

Corrections applied to this briefing

Reserved filenames corrected to index.md / log.md; the "renamed product" claim excluded as fact and flagged unverified; the "Gemini" attribution softened to "a second LLM pass"; and "folder beats vector DB" reframed as scale-bounded, citing the originator's own qmd recommendation.