Writing knowledge base

Skill auto-medica-labs/okf-cli/skills/writing-knowledge-base

Converts plain markdown into OKF-conformant knowledge bundles

Install

npx -y skills add auto-medica-labs/okf-cli --skill writing-knowledge-base

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Turn user provided content on any topic into a plain Markdown knowledge base. Invoked as /writing-knowledge-base with content to convert.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.4 KB, as published. Nobody here has run it

Knowledge Base Construction

Convert the user's content directly into a structured Markdown knowledge base. Don't ask clarifying questions unless the content is genuinely unusable (e.g. empty or unintelligible) — infer structure, terminology, and grouping from what's given, and note any gaps inline rather than blocking on them.

Workflow

Read the content and split it into coherent concepts (one idea per file — prefer several small linked files over one large one).
Create new directory with a name relative to the content user provide (e.g. italian-food, turbocharge, formula-one, etc. ).
Group concepts into type directories that fit the material's own structure. Let the content dictate the categories — e.g. recipes/, policies/, species/, procedures/, terms/, people/, events/ — whatever natural groupings emerge. Don't force a data/tech taxonomy onto non-technical content.
Write each file using the format below, preserving the user's own terminology and facts. Never invent details, numbers, names, or steps that weren't given.
Add relative Markdown links between related concepts.
Save all files under an input directory your just created.

Example file tree

cooking/
  recipes/
    italian/
      lasagna.md
      carbonara.md
    japanese/
      ramen.md
      sushi.md
  techniques/
    knife-skills.md
    emulsion.md
  ingredients/
    eggs.md
    flour-types.md
  faq.md
  common-mistake.md

Root files (e.g. faq.md) are allowed
File can be nested as deep as necessary and semantically sound.

File format

# Clear Concept Title

> One-sentence factual summary.

Body: context, details, steps, examples, caveats — whatever fits the concept.

First line # Title, then a > description, then body detail.
Use headings, lists, tables, or code blocks as they help retrieval — not all concepts need all of these.
One concept per file. Directory name per type (e.g. recipes/lasagna.md → type recipes).
Lowercase, stable filenames, no spaces.
No index.md, log.md, README.md as concepts — bundler reserves these.
No YAML frontmatter in source files

Do not fabricate any information on your own, use only what user provided to you.