New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Open Knowledge Format

by pmndrs in pmndrs/glyph

Create, migrate, inspect, query, validate, or maintain Open Knowledge Format v0.2 bundles made from linked Markdown concepts with YAML provenance.

MITAuto-check passedKnowledge Management

Install Open Knowledge Format

skills CLI
$ npx skills add pmndrs/glyph --skill open-knowledge-format -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install pmndrs/glyph open-knowledge-format --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/pmndrs/glyph.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/open-knowledge-format .claude/skills/open-knowledge-format && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
open-knowledge-format
GitHub stars
392
Token cost
~2.5k tokens
SKILL.md length
1,190 words
Files
12 (incl. scripts, references)
Skills in repo
8
Repo updated
First seen
Licence
MIT

At a glance

Create, migrate, inspect, query, validate, or maintain Open Knowledge Format v0.2 bundles made from linked Markdown concepts with YAML provenance.

  • Works in 3 steps: Every non-reserved Markdown concept has… → Every concept frontmatter block has a… → Present index.md and log.md files follow…
  • The user mentions OKF
  • SKILL.md covers Preserve the conformance…, Apply the producer profile, Choose the operation and Record changes and decisions, plus 3 more sections
  • Runs JavaScript scripts from its folder; calls node

What it does

Open Knowledge Format is an agent skill from pmndrs/glyph. Create, migrate, inspect, query, validate, or maintain Open Knowledge Format v0.2 bundles made from linked Markdown concepts with YAML provenance. Use when the user mentions OKF, Open Knowledge Format, knowledge bundles, LLM wikis, portable agent knowledge, OKF conformance, provenance, trust, lifecycle, attested computations, or asks to make repository knowledge interoperable across human and agent tools. Do not use for ordinary Markdown unless OKF compatibility or a knowledge bundle is requested.

Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 14 other files, including scripts and reference files (for example `agents/openai.yaml` and `references/okf-v0.2.md`).

It sits in Knowledge Management, covering LLM wikis and Markdown. The repository describes itself as: ♠️ A typography engine for web graphics. The licence is MIT.

When your agent uses it

  • The user mentions OKF
  • Open Knowledge Format
  • Knowledge bundles
  • Portable agent knowledge

Example prompts

  • “/open-knowledge-format”

Requirements

  • Node.js

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Every non-reserved Markdown concept has parseable YAML frontmatter.
  2. Every concept frontmatter block has a non-empty type.
  3. Present index.md and log.md files follow their reserved structures.

What it can do on your machine

Read from SKILL.md and the folder at commit 51a1f34. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 9 files in scripts/ (JavaScript), which the agent can run.

    Shell commands in SKILL.md call:

    • node

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Open Knowledge Format loads about 2.5k tokens when it runs, and up to ~4.1k if it reads all its reference files. Until then it costs about 131 tokens; SKILL.md has 1,190 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~131
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~4.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from pmndrs/glyph at commit 51a1f34, republished under its MIT licence (© pmndrs). 1,190 words, ~2,506 tokens.

Download SKILL.mdSave it as .claude/skills/open-knowledge-format/SKILL.md (or your agent's skills folder). This skill also uses 11 other files; get the full folder from GitHub.
name
open-knowledge-format
description
Create, migrate, inspect, query, validate, or maintain Open Knowledge Format v0.2 bundles made from linked Markdown concepts with YAML provenance. Use when the user mentions OKF, Open Knowledge Format, knowledge bundles, LLM wikis, portable agent knowledge, OKF conformance, provenance, trust, lifecycle, attested computations, or asks to make repository knowledge interoperable across human and agent tools. Do not use for ordinary Markdown unless OKF compatibility or a knowledge bundle is requested.

Open Knowledge Format

Apply the current upstream OKF v0.2 specification faithfully while keeping bundles useful to humans and agents. OKF is an interoperability format, not a domain taxonomy or replacement for OpenAPI, schemas, ADRs, or Diátaxis.

Read references/okf-v0.2.md completely before creating, migrating, or validating a bundle. When internet access is available and exact conformance matters, verify the current upstream specification before acting.

Preserve the conformance boundary

Establish the bundle root first. Only Markdown inside that root belongs to the bundle.

Treat exactly these as hard v0.2 requirements:

  1. Every non-reserved Markdown concept has parseable YAML frontmatter.
  2. Every concept frontmatter block has a non-empty type.
  3. Present index.md and log.md files follow their reserved structures.

Do not reject unknown types or fields, missing optional metadata, missing indexes, or broken links. Report those separately as producer-quality issues.

Apply the producer profile

For every concept authored or meaningfully changed with this skill:

  1. Require generated.by and generated.at; use the truthful v0.2 actor convention and an ISO 8601 datetime for the current content revision.
  2. Do not write the legacy v0.1 timestamp field.
  3. Put provenance in sources. Every source entry requires resource; add a stable id when a body claim uses a footnote such as [^source-id].
  4. Do not create a legacy body # Citations list. Use sources and claim-level footnotes when attribution materially improves trust.
  5. Strongly encourage resource for a concept describing a canonical asset, API, schema, dataset, package, or external system. Do not invent one for an abstract concept.
  6. Add status, stale_after, verified, credibility signals, or attestation fields only when evidence warrants them. Absence is meaningful; never fabricate trust. When status is absent, consumers treat the concept as stable.
  7. Preserve unknown producer fields.
  8. Resolve every changed local link and verify every changed external source for reachability and semantic relevance.

Choose the operation

Create or convert
  1. Identify the domain, authoritative sources, consumers, and bundle root.
  2. Inventory atomic concepts and fix their paths before cross-linking.
  3. Write one coherent concept per non-reserved Markdown file.
  4. Add type, useful descriptive metadata, truthful generated, and warranted sources.
  5. Add concise indexes for progressive disclosure and a log only when useful.
  6. Verify links and sources, then validate hard conformance separately from producer-profile errors and warnings.
Migrate v0.1 to v0.2
  1. Read §13 of the current specification.
  2. Change the root declaration to okf_version: "0.2".
  3. Replace timestamp with truthful generated.by and generated.at.
  4. Move final # Citations entries into sources; split entries containing multiple links into separate sources and retain their titles.
  5. Convert logs to one H1 title followed by newest-first ## YYYY-MM-DD sections.
  6. Preserve all other fields and prose, then remove the legacy fields and citation section.
  7. Run node scripts/validate-okf.mjs <bundle-root> from this skill directory and resolve every error.
Maintain
  1. Inspect changed sources and affected concepts.
  2. Update facts and relationships without deleting unknown fields.
  3. Refresh generated.by and generated.at for meaningful content edits.
  4. Update sources and claim footnotes when provenance changes.
  5. Update relevant indexes and record the change as one new Log Entry (see Record changes and decisions).
  6. Reverify affected links, sources, and fragments.
  7. Validate and report hard errors, producer-profile errors, and warnings separately.
Query
  1. Start at the root index.md, otherwise inventory paths and frontmatter.
  2. Use type, title, description, tags, sources, status, trust, lifecycle, and links to select concepts.
  3. Locate before reading. node scripts/docs-query.mjs <bundle> search <terms…> prints each matching paragraph under path › Heading › Subheading [start-end]; a single path-like term (src/three/text.ts) also lists the concepts that cite or link that file. outline <path> prints the heading tree with line ranges, outline <path>:<line> the sections containing a line, and decision <D-n|slug> one decision. Then read only the reported range.
  4. Read only the bodies required to answer.
  5. Cite concept paths used and distinguish bundle facts from inference.
Show full SKILL.md (551 more words)Show less
Validate

Run the bundled validator with Node.js:

sh
node scripts/validate-okf.mjs /path/to/bundle

For a repository that maintains Workspace Package concepts, require complete package coverage and source freshness:

sh
node scripts/validate-okf.mjs /path/to/bundle --workspace-root /path/to/repository
node scripts/docs-drift.mjs /path/to/repository --markdown drift.md          # repository-wide drift issue body
node scripts/docs-drift.mjs /path/to/repository --pr origin/main --markdown r.md  # advisory pull-request report

The validator discovers apps/*/package.json, benches/package.json, and packages/*/package.json. Each manifest requires exactly one type: Workspace Package concept whose workspace_package and resource match; a retired source_digest field is a producer-profile error. Freshness is not stored in a concept, because a stored source hash conflicts on every concurrent change to the same package. Instead attestations.mjs records intent and verification as new files: attest <package> <note> writes an Attestation naming the package source as a SHA-256 over git blob IDs; verify <slug> writes a verification Log Entry with verified_sources and a reviews row (kind attestation or gap, verdict confirmed, corrected, documented, or no-change) and removes the consumed attestations; audit lists pending attestations and gaps since the last verification. docs-drift.mjs renders the advisory pull-request report (--pr <base> --head <sha>) and the tracking-issue body. Audits need full git history and refuse a shallow clone.

Report:

  • Conformance errors: violations of the three hard requirements.
  • Producer-profile errors: legacy v0.1 fields, missing or malformed generated, malformed source families, invalid or unverified links/sources, or unjustified trust fields.
  • Warnings: missing recommended metadata, weak navigation, orphan concepts, indirect sources, or potentially stale claims.

Record changes and decisions

Records are append-only and never share a file, so concurrent changes cannot conflict over them. Create each one with the scaffolder, which refuses to overwrite an existing subject, then replace every TODO(docs:new); validation rejects leftover scaffold text.

sh
node scripts/records.mjs <bundle> new log <slug> <title…>
node scripts/records.mjs <bundle> new decision <slug> <title…>
node scripts/records.mjs <bundle> list log [--since YYYY-MM-DD] [--mentions <text>] [--limit n | --all]
node scripts/records.mjs <bundle> list decision
  • A Log Entry concept lives at log/YYYY-MM-DD-<slug>.md: title and generated in frontmatter, flat prose in the body, no headings.
  • A Decision concept lives at …/decisions/<slug>.md, named by its subject with no number prefix. It carries decision_status (Proposed, Experiment, Deferred, Accepted, or Superseded), a quoted decided date, optional supersedes (register IDs or decision slugs), and its rule under ## Decision, followed by ## Why and ## Consequences.
  • Never number a record from a shared counter, and never add to a shared index; list derives the newest-first view.
  • A legacy register that declares frozen_after: D-<n> accepts no row numbered above it. Existing rows keep their IDs and may still change status.
  • A bundle that records changes as Log Entry files has no log.md; validation flags one that reappears.

Handle reserved files

  • Root index.md may contain only okf_version: "0.2" in frontmatter.
  • Nested indexes have no frontmatter and provide concise navigation.
  • Logs have one H1 title and newest-first ## YYYY-MM-DD sections with flat prose entries. A bundle may instead record each change as a Log Entry file under log/ and omit log.md.
  • Never treat index.md or log.md as concepts.

Handle provenance and trust

  • Use sources[].resource for the material a concept derives from.
  • Use the actor forms <producer>/<version>, human:<id>, and process:<id> for generated.by and verified[].by exactly as specified.
  • Treat verified as by/at verification history, not a confidence score; consumers accept a bare mapping as a one-item list.
  • Derive trust tiers and freshness from the standard fields. Do not store a subjective credibility score.
  • Keep each attested computation as its own concept and never let an agent rewrite its sanctioned computation during execution.

Compose with Diátaxis

Use OKF for portable structure, provenance, trust, and links. Use Diátaxis to decide whether reader-facing material is a tutorial, how-to, reference, or explanation. Keep internal plans, decisions, and schemas in their native formats while representing each as a coherent OKF concept.

© pmndrs, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 11 other files (scripts, references) in .agents/skills/open-knowledge-format of pmndrs/glyph.

  • SKILL.md
  • agents/openai.yaml
  • references/okf-v0.2.md
  • scripts/attestations.mjs
  • scripts/attestations.test.mjs
  • scripts/docs-drift.mjs
  • scripts/docs-query.mjs
  • scripts/migrate-v01-to-v02.mjs
  • scripts/okf-tools.test.mjs
  • scripts/records.mjs
  • scripts/validate-okf.mjs
  • scripts/workspace-packages.mjs

Open the folder on GitHubat commit 51a1f34

Compare with similar skills

Open Knowledge Format next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Open Knowledge Format compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Open Knowledge Format this skillpmndrs/glyph392—~2.5kAutomated safety check: PassMIT
Okf Open Knowledge Formatfabricioctelles/skills105—~5.7kAutomated safety check: PassApache-2.0
Baoyu URL To Markdownsdyckjq-lab/llm-wiki-skill2.5k2 repos~3.2kAutomated safety check: PassNone
Karpathy LLM WikiAstro-Han/karpathy-llm-wiki2.4k—~3.6kAutomated safety check: PassMIT
LLM Wiki Knowledge GraphEgonex-AI/Understand-Anything85k1 repos~1.5kAutomated safety check: PassMIT
Hermes History IngestAr9av/obsidian-wiki3.5k1 repos~2.2kAutomated safety check: NotesMIT

Similar skills

  • Okf Open Knowledge Format

    fabricioctelles/skills

    Create, validate, and enrich Open Knowledge Format (OKF) bundles — the open spec for representing organizational knowledge as markdown files with YAML frontmatter.

    105 GitHub stars~5.7k tokensUpdated 3 days ago
    Knowledge ManagementAuto-check passed

  • Baoyu URL To Markdown

    sdyckjq-lab/llm-wiki-skill

    Fetch any URL and convert to markdown using Chrome CDP. An agent skill from sdyckjq-lab/llm-wiki-skill.

    2.5k GitHub starsUsed in 2 repos~3.2k tokens
    Knowledge ManagementAuto-check passed

  • Karpathy LLM Wiki

    Astro-Han/karpathy-llm-wiki

    A skill your agent uses when building or maintaining a personal LLM-powered knowledge base.

    2.4k GitHub stars~3.6k tokensUpdated 2 mo ago
    Knowledge ManagementAuto-check passed

  • LLM Wiki Knowledge Graph

    Egonex-AI/Understand-Anything

    Detects a Karpathy-pattern LLM wiki and builds an interactive knowledge graph with entities, implicit relationships and topic clusters.

    85k GitHub starsUsed in 1 repo~1.5k tokens
    Knowledge ManagementAuto-check passed

  • Hermes History Ingest

    Ar9av/obsidian-wiki

    Ingest Hermes agent history into Obsidian as distilled knowledge.

    3.5k GitHub starsUsed in 1 repo~2.2k tokens
    Knowledge ManagementAuto-check: notes

  • LLM Wiki

    lewislulu/llm-wiki-skill

    Build and maintain a Karpathy-style LLM knowledge base — a self-compiling Obsidian markdown wiki where an Agent ingests raw sources, compiles cross-linked concept/entity/summary pages, answers…

    655 GitHub stars~3.7k tokensUpdated 5 mo ago
    Knowledge ManagementAuto-check passed

More from pmndrs/glyph

All 8 skills in this repo

  • Codemod

    pmndrs/glyph

    Author, archive, apply, and verify TypeScript codemods with ts-morph for APIs that have reached the remote default branch or external users.

    392 GitHub stars~1.2k tokensUpdated today
    Auto-check passed

  • Tsl

    pmndrs/glyph

    Implement, migrate, review, debug, or verify Three.js Shading Language (TSL) materials, node graphs, WebGPU compute work, and post-processing.

    392 GitHub stars~1.5k tokensUpdated today
    Auto-check passed

  • Diataxis Docs

    pmndrs/glyph

    Design, classify, write, audit, or restructure technical documentation with the Diátaxis framework.

    392 GitHub stars~1.4k tokensUpdated today
    Auto-check passed

  • Maintainability Review

    pmndrs/glyph

    Audit and improve repository code milestone by milestone for correctness, clarity, local reasoning, DRY design, explicit state modeling, panic resistance, and trustworthy TypeScript boundaries.

    392 GitHub stars~1.7k tokensUpdated today
    Auto-check passed

  • Evidence First

    pmndrs/glyph

    Shape human-facing engineering communication—including chat updates and final answers, reports, reviews, handoffs, debugging or benchmark summaries, PR and issue prose, READMEs, and technical…

    392 GitHub stars~1.2k tokensUpdated today
    Auto-check passed

  • Gh Stack

    pmndrs/glyph

    Manage dependent branches and pull requests with the gh-stack GitHub CLI extension.

    392 GitHub stars~1.1k tokensUpdated today
    Auto-check passed

Categories

Questions about Open Knowledge Format

What does Open Knowledge Format do?

Create, migrate, inspect, query, validate, or maintain Open Knowledge Format v0.2 bundles made from linked Markdown concepts with YAML provenance. Open Knowledge Format is an agent skill from pmndrs/glyph.2 bundles made from linked Markdown concepts with YAML provenance.

When should I use Open Knowledge Format?

Open Knowledge Format fits situations like: the user mentions OKF; open Knowledge Format; knowledge bundles; portable agent knowledge.

How do I install Open Knowledge Format in Claude Code?

Run `npx skills add pmndrs/glyph --skill open-knowledge-format -a claude-code`. Or copy the skill folder (.agents/skills/open-knowledge-format in pmndrs/glyph) into .claude/skills/open-knowledge-format in your project. Claude Code loads it when a task matches its description.

How do I install Open Knowledge Format in Codex?

Run `npx skills add pmndrs/glyph --skill open-knowledge-format -a codex`. Or copy the skill folder (.agents/skills/open-knowledge-format in pmndrs/glyph) into .agents/skills/open-knowledge-format in your project. Codex loads it when a task matches its description.

Can I use Open Knowledge Format in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add pmndrs/glyph --skill open-knowledge-format -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/open-knowledge-format, .gemini/skills/open-knowledge-format, .github/skills/open-knowledge-format and .opencode/skills/open-knowledge-format in your project.

What does Open Knowledge Format need to run?

Going by SKILL.md and its folder, Open Knowledge Format needs JavaScript for the scripts in its folder and the command-line tools its instructions call (node). Our summary lists: Node.js.

Does Open Knowledge Format access the network?

SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.

Is Open Knowledge Format safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Open Knowledge Format use?

Open Knowledge Format is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Open Knowledge Format use?

About 2.5k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.6k tokens, read only when the agent opens those files.

What are the alternatives to Open Knowledge Format?

Skills that share tags, products or a category with Open Knowledge Format: Okf Open Knowledge Format (fabricioctelles/skills, 105 stars), Baoyu URL To Markdown (sdyckjq-lab/llm-wiki-skill, 2.5k stars), Karpathy LLM Wiki (Astro-Han/karpathy-llm-wiki, 2.4k stars) and LLM Wiki Knowledge Graph (Egonex-AI/Understand-Anything, 85k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Open Knowledge Format?

pmndrs (a GitHub organization) maintains it in pmndrs/glyph, which has 392 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on October 7, 2026.

Source: pmndrs/glyph on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.