LCK Analytics Skill (NomaDamas) for Claude Code: What It Does and How to Install It
Makes Claude look up LCK League of Legends results and standings, and analyse live games, drafts and patch meta.
- Installs
- 8.9K
- on skills.sh
- Stars
- 8.2K
- on GitHub
- Files
- 10
- besides SKILL.md
- Updated
- Nov 2026
- last repo commit
It answers prompts like "오늘 LCK 경기 결과 알려줘" or "지금 T1 경기 킬/골드/오브젝트 요약해줘". It reads results, standings and live stats from Riot's official LoL Esports surfaces, and builds team power ratings, champion matchup and synergy views and patch meta from Oracle's Elixir style historical data you supply. Three bundled Node scripts handle the local pipeline: sync-oracle.js, build-match-report.js and analyze-live-game.js.
SKILL.md is a stub that has Claude run an npx command to print the live instructions. instruction.md then says to install the lck-analytics npm package globally first, and not to work around a missing package. The skill is for lookup and analysis only, and its legal disclaimer limits use to personal, non-organizational lookups. It was moved into k-skill from jerjangmin's original skill pack.
Good for
- LCK fans who want results and standings for a date or team
- Analysts who have Oracle's Elixir style CSV data and want power ratings, matchups and patch meta
- Developers who want live game stats and turning points from a script
Install lck-analytics
Claude Code
- 1
Copy this prompt.
Install the lck-analytics skill in Claude Code. Run this command: /plugin marketplace add NomaDamas/k-skill /plugin install k-skill@k-skill Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js 18 or newer.
Pick one route. The plugin route installs the k-skill plugin, which covers the whole repo: 135 top-level skill folders (counted from the repo tree), with no hooks or MCP config files in the tree. The skills CLI and folder routes install only lck-analytics. The skill also needs a global npm install of lck-analytics. The skill runs npx at use time, so there is no Claude app upload route.
From the project README. Source.
What's in the skill
When it loads
Riot 공식 LoL Esports 데이터와 Oracle's Elixir 스타일 historical 데이터로 LCK 경기 결과, 현재 순위, live turning point, 밴픽 matchup/synergy, patch meta, 팀 파워 레이팅을 조회한다.
Sections in SKILL.md
- Get the full instructions (required first step)
- Legal disclaimer (required)
- Hard rules even without the CLI
Key facts
- License
- MIT
- Installs on skills.sh
- 8,876 (Nov 2, 2026)
- GitHub stars
- 8,232 for the whole repo (Nov 2, 2026)
- Last repo commit
- Nov 1, 2026
- Audits on skills.sh
- Gen Agent Trust Hub: warn, Socket: pass, Snyk: pass
What it includes
SKILL.md tells Claude to run this first and follow its output
Fetches the current instructions from the k-skill npm package. The repo copy is lck-analytics/instruction.md.
Runs code ·
Bootstrap: npx -y @nomadamas/k-skill@0 instruct lck-analyticsinstruction.md lists a global install as a prerequisite
It also says to try this first instead of working around a missing package.
Runs code ·
Install the package: npm install -g lck-analyticsgetLckSummary(date, { team, includeStandings }) for a YYYY-MM-DD date
Team aliases such as 한화, 젠지 and T1 are normalised.
Try: “2026-04-01 한화 경기 결과랑 순위 보여줘”
Runs code ·
Match results and standingsscripts/analyze-live-game.js --game <game id>
Uses live window and details feeds from feed.lolesports.com, or fixture files passed with --window and --details.
Try: “이 경기 turning point가 뭐였어?”
Runs code ·
Live game stats and turning pointsscripts/sync-oracle.js --csv <file> builds a historical cache JSON
The default cache folder is .openclaw-lck-cache/.
Runs code ·
Sync Oracle-style CSVscripts/build-match-report.js --date <YYYY-MM-DD> [--team <name>]
Optionally filtered to one team.
Runs code ·
Match report by dateTeam power ratings, champion matchup and synergy, patch meta from historical rows
instruction.md says to show the sample size first and mark small samples as low confidence.
Try: “현재 패치에서 어떤 챔피언이 메타 픽인지 보여줘”
Read ·
Ratings, matchups and patch meta
From the official docs
Permissions
High risk
Claude runs npm packages and Node scripts, follows instructions fetched at run time and writes cache files. Read the output of the k-skill command before Claude acts on it, and check the lck-analytics package before the global install.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
Claude runs whatever the k-skill npm package prints
SKILL.md says to run npx -y @nomadamas/k-skill@0 instruct lck-analytics and follow its output as the primary instructions. Claude therefore acts on text that comes from an npm package at run time, and the @0 tag floats across 0.x releases, so the instructions can change without any change to SKILL.md. Read the output before Claude acts on it. github.com
npx -y installs packages without asking
The bootstrap and update commands use npx -y, which downloads and runs the k-skill package from npm without a prompt. github.com
It installs a global npm package
instruction.md lists npm install -g lck-analytics as a prerequisite and says to try it before any other route. Check the package on npm before Claude installs it. github.com
It runs bundled scripts that write cache files
sync-oracle.js, build-match-report.js and analyze-live-game.js create folders and write JSON under .openclaw-lck-cache/ in the current working directory by default (the --cache option changes it), and _lib.js calls npm root -g to find the global package. github.com
The k-skill CLI sends usage statistics
The docs say the CLI can send a usage event to PostHog with a random local id, the command, the skill name, CLI version, runtime mode and a success flag. It does not send command arguments, input, file contents, API responses or credentials. Set KSKILL_ANALYTICS_DISABLED=1 to opt out. github.com
The plugin route installs the whole k-skill set
The k-skill plugin in the repo tree has 135 top-level skill folders, including skills that cover other services. The skills CLI route can install just this one. github.com
Legal disclaimer: personal lookups only
SKILL.md says not to use the skill for systematic or bulk crawling, database building or circumventing blocks, and links a full Korean disclaimer in references/DISCLAIMER.md. github.com
Hard rules in SKILL.md
SKILL.md says never to run payment, message delivery, final submission, cancellation or public posting without explicit approval, and never to handle plaintext credentials in chat or files. github.com
Security advisories
No published security advisories as of Nov 2, 2026.
Technical details
| Turning on read-only | No read-only switch exists. |
|---|---|
| Talks to | feed.lolesports.com (live game feeds named in instruction.md). Riot LoL Esports schedule, standings and event endpoints used by the lck-analytics package. npm registry (k-skill and lck-analytics packages). us.i.posthog.com (k-skill usage statistics, opt-out available) |
| Sign-in and keys | SKILL.md says never to ask for or store plaintext credentials. |
| Scopes |
|
Questions
What does the lck-analytics skill do?
It makes Claude answer LCK questions: match results for a date, current standings, live game stats and turning points. With Oracle's Elixir style data you supply, it can also rate teams and summarize matchups and patch meta.
How do I install it?
Use the install box at the top of this page. The repo README documents a Claude Code plugin for the whole k-skill set and a skills CLI command for one skill. The skill also needs Node.js 18 or newer and a global npm install of lck-analytics.
Do I need an API key?
No key from you is listed. instruction.md names a LOLESPORTS_API_KEY public fallback and says an environment variable override may be needed if that fallback is rotated.
Is it an official Riot or LCK product?
No. SKILL.md says the skill is not an official feature of, supported by or affiliated with any trademark owner it names. It uses Riot official data surfaces, which can change and then need a package fix.
Is it safe?
It has Claude run npx -y and a global npm install, which download and run third-party code, and it follows instructions fetched at run time. The bundled scripts write cache files, by default under .openclaw-lck-cache/. Read what the k-skill command prints before Claude acts on it.
Can I crawl LCK data in bulk with it?
No. The legal disclaimer says automated collection of public information must stay personal and non-organizational, with no systematic or bulk crawling, database building or circumvention of blocks.
Who made it?
The README says the skill pack was created by jerjangmin and moved into the k-skill repo, with packaging changes for its npm workspace and release flow.
Information
updated Nov 2, 2026
- Repository
- nomadamas/k-skill
- Folder
lck-analytics
Sources
Docs, issue trackers and discussions this page is built from.
Show all 7Hide
- k-skill README · github.com
- k-skill marketplace.json · github.com
- lck-analytics SKILL.md · github.com
- lck-analytics instruction.md · github.com
- lck-analytics README.md · github.com
- _lib.js · github.com
- k-skill usage analytics · github.com