Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/roboco-io/plugins/llm-wikinpx skills add roboco-io/plugins --skill llm-wikigit clone --depth 1 https://github.com/roboco-io/pluginsWhat 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 | $0.00129 | $0.02346 |
| Opus 5 | $0.00064 | $0.01173 |
| Sonnet 5 | $0.00026 | $0.00469 |
| Haiku 4.5 | $0.00013 | $0.00235 |
Grade A, and why
llm-wiki 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 2d 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 — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.
llm-wiki — 프로젝트 컨텍스트 위키
참고: Karpathy LLM Wiki. 검색이 아닌 컴파일이 핵심 — 한 번 이해한 것은 계속 다듬어 재활용한다.
언제 사용하나
트리거: "위키 초기화", "wiki 초기화", "wiki ingest", "문서 wiki로 정리", "프로젝트 컨텍스트 위키", "llm wiki", "knowledge base 구축", "wiki query", "서드파티 소스 컴파일".
이 스킬은 워크스페이스 CLAUDE.md의 "LLM Wiki 활용 규칙"을 실제 실행 가능하게 한다.
디렉토리 구조
프로젝트 루트에 init 명령으로 생성되는 구조:
wiki/
├── index.md # 자동 재빌드 (수동 편집 금지)
├── log.md # ingest 이력
├── raw/ # 원본 소스 드롭존 (불변)
│ └── YYYY-MM-DD_{source}.md
├── {page}.md # 컴파일된 위키 페이지 ([[wiki-link]] 교차참조)
├── .gitignore # raw/private/, .lancedb/ 기본 제외
└── .lancedb/ # 선택적 벡터 인덱스 (lancedb-sync 시)
프로젝트 루트 CLAUDE.md에는 "LLM Wiki 활용 규칙" 블록이 init 시 자동 추가된다.
페이지 프론트매터 규약
---
title: "Rate Limiter Design"
type: decision # concept | decision | entity | meeting | policy | comparison
tags: [api, reliability]
status: active # draft | active | stale | archived
confidence: high # high | medium | low | tentative
created: 2026-04-19
updated: 2026-04-19
sources: [raw/2026-04-18_arch-review.md]
---
상세는 page-templates.md 참조.
8개 명령
init — 구조 초기화
실행: scripts/init.sh [project-root]
wiki/디렉토리 구조 생성 (raw/포함)wiki/index.md,log.md,.gitignore초기 파일- 프로젝트 루트
CLAUDE.md에 LLM Wiki 블록 추가 (claude-md-block.md) - qmd / LanceDB 환경 감지 및 설치 안내
ingest <source> — 소스 컴파일
실행: LLM 주도 워크플로 (스크립트 없음)
상세 5단계: ingest-workflow.md
- 소스 읽기 (
raw/에 드롭된 파일 또는 인자 경로) - 엔티티·개념 추출
- 기존 wiki 페이지와 교차 참조 검토
- 페이지 생성/갱신 (10-15 페이지)
log.md에 기록
페이지 유형 선택은 page-templates.md 6종(concept/decision/entity/meeting/policy/comparison) 참조.
query <질문> — 위키 검색
qmd 감지 시 (하이브리드 검색):
qmd query "<질문>"호출- BM25 + 벡터 + LLM 재순위 결과 top-10
- LLM이 컨텍스트 트리 + 매칭 페이지로 답변 합성
- 답변에
[[wiki-link]]로 출처 인용
What ships with it
11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/claude-md-block.md 1.7 KB
- references/ingest-workflow.md 4.5 KB
- references/karpathy-pattern.md 3.7 KB
- references/lancedb-integration.md 3.3 KB
- references/page-templates.md 5.3 KB
- references/qmd-integration.md 2.5 KB
- scripts/init.sh 2.5 KB runs code
- scripts/lancedb_sync.py 3.2 KB runs code
- scripts/lint_wiki.py 3.3 KB runs code
- scripts/qmd_index.sh 780 B runs code
- scripts/sync_index.py 3.8 KB runs code
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.
- 2d ago First seen · 187 lines · 129 tokens per session scan A b41c9918c7c4
llm-wiki is a skill published in the GitHub repository roboco-io/plugins (21 stars, last pushed 1mo ago), licensed MIT. It adds 129 tokens to every session and 2,346 once invoked, about $0.0006 per session on Opus 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.