Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/kimsanguine/llm-brainnpx agentmods add commands/kimsanguine/llm-brain/okfWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/kimsanguine/llm-brain/okf)<a href="https://agentmods.dev/commands/kimsanguine/llm-brain/okf"><img src="https://agentmods.dev/badge/commands/kimsanguine/llm-brain/okf/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/kimsanguine/llm-brain/okf"><img src="https://agentmods.dev/badge/commands/kimsanguine/llm-brain/okf.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00014 | $0.01408 |
| Opus 5.5 | $0.00006 | $0.00563 |
| Sonnet 5.5 | $0.00003 | $0.00282 |
| Haiku 4.5 | $0.00001 | $0.00141 |
Grade A, and why
okf scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
llm-brain의 okf 커맨드입니다. wiki/(내부 슈퍼셋)를 OKF v0.1 호환 번들 okf/로
투영합니다. 기본 경로는 기존 개인용 export이고, 외부 공유는 별도의 --share gate만 사용합니다.
$ARGUMENTS에 따라 아래를 실행하세요.
인자 파싱
--dry-run→ 파일 미작성, export 대상·제외·통계만 출력 (보안 검토용)--strip-internal→ OKF 예약 6필드만 (x-llmbrain-* 제거, 단독으로는 공유 승인 아님)--exclude-slug <slug>→ 특정 페이지 추가 제외 (복수 가능)--config <경로>→ private export 전용 override.--share는 canonicalschema/okf_export.yaml만 허용--out <경로>→ 출력 번들 루트 (기본 privateokf/, shareokf-share/)--share→ Share-ready gate + redacted manifest를 거쳐 별도 공개 번들 생성--approve-share I_ACKNOWLEDGE_SHARE_READY_EXPORT→ 사람이 직접 입력하는 필수 승인 값- 인자 없음 → 기본 export (
okf/생성)
기본 private export
기존 uv run python scripts/okf_export.py [--dry-run] 동작은 호환 유지한다. 이는 개인용
OKF projection이며 Share-ready 승인을 의미하지 않는다. --strip-internal만으로도 공개 승인이
되지는 않는다.
🔴 Share-ready gate (외부 공유 전 one-way door)
공유 산출물을 push하면 history가 영구다(비가역). 민감정보 누출을 막는다:
schema/okf_export.local.yaml(gitignored) 존재 확인. 아래 두 키를 명시한다. 민감 키워드(sensitive_patterns)· 제외 페이지(exclude_slugs)는 반드시 이 로컬 파일에만 둔다. 커밋되는schema/okf_export.yaml엔 실명·내부명을 절대 넣지 않는다(그 자체가 누출).- 공유 후보 페이지는
scope: shared|private가 반드시 있어야 한다. 미지정·알 수 없는 scope, 허용되지 않은 classification(type), skipped/broken page는 hard-stop이다. - 다음 명령의 승인 값은 자동 생성·추론하지 말고 사람이 검토 후 직접 입력한다.
exclude_slugs: []
sensitive_patterns: []
uv run python scripts/okf_export.py --share \
--approve-share I_ACKNOWLEDGE_SHARE_READY_EXPORT
게이트는 base policy 또는 local security config 부재, 민감 hit, scope/policy 위반, wiki 내부
symlink/외부 해석 경로면 출력 전에 중단한다. local config는 exclude_slugs와
sensitive_patterns만 추가할 수 있고 base policy·승인 값을 바꿀 수 없다.
--out의 canonical path가 레포 루트, raw/**, wiki/** 안이면 stage·receipt·output을
만들기 전에 거부한다.
통과 시 sibling stage에 전체 bundle과 share-manifest.json을 만든다. 기존 디렉토리 교체는
portable POSIX에서 단일 atomic rename이 아니므로 두 rename 사이 아주 짧게 경로가 없을 수 있다.
첫 rename 전에 durable recovery receipt를 만들며, hard exit 뒤에는 다음 정상 share 실행이 기존
bundle 복구 또는 새 bundle 완료를 수행한다. 따라서 published bundle과 recovery receipt가 동시에
없는 상태는 만들지 않는다.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 100 lines · 14 tokens per session scan A 7cdbcee0512b
okf is a command published in the GitHub repository kimsanguine/llm-brain (10 stars, last pushed 21d ago), licensed MIT. It adds 14 tokens to every session and 1,408 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
ingest
Ingest a source (URL/file/text) into Raw Sources + compile 1015 Wiki pages, with mandatory user-purpose gate and mothership cross-linking.
lint
Run comprehensive wiki health check — orphans, broken links, contradictions, stale pages, index sync, MOC coverage, v2/v4/v5 frontmatter coverage, Core Context freshness, and cross-vault link integrity (mainVaultRelated/mainVaultCmds).
capture-tabs
Capture a Chrome research tab group or AI-chat research session into LLM Wiki Inbox markdown, then optionally route it through /inbox or /ingest.
audit
Audit the whole Wiki vault against 3 knowledge-integrity criteria — eligibility coverage, MOC-cluster consistency, confidence calibration. Produces a vault health report and queues high-priority pages for /verify.
verify
Verify a single Wiki page against 3 knowledge-integrity criteria — eligibility, consistency, confirmability. Writes verificationStatus back to the page; flags conflicts as disputed rather than deleting them.
onboard
An interview-based setup command for an LLM Wiki, a notes system designed for use with language models. It asks about your wiki, files, working mode, and personal context, then fills in the setup details.