Ingest into an OKF knowledge base
Add a source or concept to an OKF wiki as one Markdown file (a "concept"). If the repo has a
CLAUDE.md, that is authoritative — follow it (this skill summarizes the same rules so it
works standalone).
0. Orient
- The bundle is the
wiki/ directory. Edit knowledge only there. A concept's id is its path
minus .md (wiki/papers/foo.md → papers/foo).
- If
tools/okf_lint.py exists, you'll run it at the end.
1. Verify — never fabricate
Fetch the real source and confirm its facts from the source itself, not memory (your training
cutoff may predate it):
- arXiv:
curl -sL "http://export.arxiv.org/api/query?id_list=<id>" — read the canonical
<title>, <author> list, and <published> date. If the id is unknown, search:
...query?search_query=ti:%22<title words>%22&max_results=5.
- Web: use WebFetch on the URL; capture title, author/site, date.
- If you cannot confirm it exists, stop and tell the user — do not invent metadata.
- Optionally save the raw capture under
raw/ (read-only provenance, outside the bundle).
2. Place it
- Choose a domain folder by subject (
wiki/papers/, wiki/people/, …) — folder = subject,
type = kind. Create the folder + an index.md if it's new.
- Choose a lowercase, hyphenated slug (matches
[A-Za-z0-9_][A-Za-z0-9_.-]*, no spaces).
3. Write wiki/<domain>/<slug>.md
Frontmatter (keys in this order; use templates/<type>.md as a stamp if present):
---
type: Source # Source | Entity | Concept | Note | Overview | Reference
resource: <URI> # required for Source
title: <human title>
description: <one sentence>
tags: [a, b] # a YAML LIST, never a comma string
timestamp: 2026-01-01T00:00:00Z # UTC ISO 8601; refresh on every edit
# extensions allowed after the standard keys (authors, year, venue, doi, …)
---
Body: grounded summary prose, then optional # Key contributions / # Schema / # Examples
(fenced code is fine), then # Citations (the source's own resource URI first). Ground every
claim in the source.
4. Link, index, log
- Add genuine relative markdown cross-links to/from related concepts —
[title](../other/x.md). No [[wikilinks]], no leading /.
- Update the domain
index.md (* [Title](slug.md) - description) and the root wiki/index.md
if you added a new domain.
- Prepend a
wiki/log.md entry under today's date (* **Creation**: Added [Title](path).).
5. Lint
python3 tools/okf_lint.py # fix any errors; review warnings
Scaling up
For many sources, or anything post-dating your knowledge cutoff, fan out with a workflow:
discover → verify each against arXiv/web → author. Verification-before-authoring is what
keeps the KB free of hallucinated sources.
1---2name: okf-ingest3description: Ingest a source or new concept into an Open Knowledge Format (OKF) knowledge base (a wiki/ bundle of Markdown + YAML frontmatter). Use when the user wants to add a paper, article, URL, video, repo, dataset, person, project, or note to their wiki / knowledge base / second brain. Triggers: "ingest this", "add this paper/article to the wiki", "capture this source", "save this to my KB", "add a note about X". Fetches and VERIFIES the source, writes a conformant page, cross-links it, updates the index + changelog, and lints.4---56# Ingest into an OKF knowledge base78Add a source or concept to an OKF wiki as one Markdown file (a "concept"). If the repo has a9`CLAUDE.md`, **that is authoritative — follow it** (this skill summarizes the same rules so it10works standalone).1112## 0. Orient1314- The bundle is the `wiki/` directory. Edit knowledge only there. A concept's id is its path15 minus `.md` (`wiki/papers/foo.md` → `papers/foo`).16- If `tools/okf_lint.py` exists, you'll run it at the end.1718## 1. Verify — never fabricate1920Fetch the real source and confirm its facts from the source itself, not memory (your training21cutoff may predate it):2223- **arXiv**: `curl -sL "http://export.arxiv.org/api/query?id_list=<id>"` — read the canonical24 `<title>`, `<author>` list, and `<published>` date. If the id is unknown, search:25 `...query?search_query=ti:%22<title words>%22&max_results=5`.26- **Web**: use WebFetch on the URL; capture title, author/site, date.27- If you cannot confirm it exists, **stop and tell the user** — do not invent metadata.28- Optionally save the raw capture under `raw/` (read-only provenance, outside the bundle).2930## 2. Place it3132- Choose a **domain folder** by subject (`wiki/papers/`, `wiki/people/`, …) — folder = subject,33 `type` = kind. Create the folder + an `index.md` if it's new.34- Choose a lowercase, hyphenated **slug** (matches `[A-Za-z0-9_][A-Za-z0-9_.-]*`, no spaces).3536## 3. Write `wiki/<domain>/<slug>.md`3738Frontmatter (keys in this order; use `templates/<type>.md` as a stamp if present):3940```yaml41---42type: Source # Source | Entity | Concept | Note | Overview | Reference43resource: <URI> # required for Source44title: <human title>45description: <one sentence>46tags: [a, b] # a YAML LIST, never a comma string47timestamp: 2026-01-01T00:00:00Z # UTC ISO 8601; refresh on every edit48# extensions allowed after the standard keys (authors, year, venue, doi, …)49---50```5152Body: grounded summary prose, then optional `# Key contributions` / `# Schema` / `# Examples`53(fenced code is fine), then `# Citations` (the source's own `resource` URI first). Ground every54claim in the source.5556## 4. Link, index, log5758- Add genuine **relative markdown** cross-links to/from related concepts —59 `[title](../other/x.md)`. **No `[[wikilinks]]`, no leading `/`.**60- Update the domain `index.md` (`* [Title](slug.md) - description`) and the root `wiki/index.md`61 if you added a new domain.62- Prepend a `wiki/log.md` entry under today's date (`* **Creation**: Added [Title](path).`).6364## 5. Lint6566```bash67python3 tools/okf_lint.py # fix any errors; review warnings68```6970## Scaling up7172For many sources, or anything post-dating your knowledge cutoff, fan out with a workflow:73**discover → verify each against arXiv/web → author**. Verification-before-authoring is what74keeps the KB free of hallucinated sources.
thattimc