
Stockmatrix Mcp
io.github.MongLong0214由 GitHub 帳號 MongLong0214 發布v0.5.0更新於 Oct 24, 2026
Korean stock market theme lifecycle analysis (TLI) — rankings, stock lookup, predictions & prompts
概覽
讓助理依據 TLI 生命週期評分分析韓國股市投資主題,提供排名、個股查詢、主題比較與預測。
- 可以做什麼
- 透過 StockMatrix API 提供 11 個工具進行韓國市場主題分析:依生命週期階段取得主題排名、AI 最佳化的市場總覽、每日或每週評分變動與階段轉換、2 至 5 個主題的並列比較,以及基於歷史相似比對的上漲、見頂、降溫預測。也可依關鍵字、公司名稱或六位代碼搜尋主題與個股,進行個股到主題的反向查詢,回傳詳細評分組成、30 天評分歷史與 TLI 方法論文件。另有 4 個工作流程提示詞打包多工具分析,2 個唯讀資源提供方法論與當日排名。
- 適用情境
- 適合想以對話方式了解韓國股市主題而非原始行情的人:追蹤哪些 KOSPI/KOSDAQ 主題正在興起或見頂、檢查持股是否屬於熱門主題、比較不同主題,或回顧主題近一個月的動能。
- 執行需求
- 以 npm 套件 stockmatrix-mcp 透過 stdio 在本機執行,使用 npx 啟動,因此需要 Node.js 以及連線 StockMatrix API 的網路。未宣告需要身分驗證。選用環境變數 STOCKMATRIX_API_URL 可覆寫預設 API 基礎網址。文件鎖定 Claude Desktop、Cursor、Windsurf、VS Code 等桌面 MCP 用戶端。
安裝
在 SourceWeft 中
- 開啟 儀表板中的 Stockmatrix Mcp,將其新增到工作區。
- 為需要使用其工具的對話啟用該服務。
Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。
其他 MCP 客戶端
參照 儲存庫 中的啟動說明。
README
stockmatrix-mcp
Official MCP server for StockMatrix (https://stockmatrix.co.kr).
Ask about Korean stock market themes in natural conversation with AI. Track 200+ KOSPI/KOSDAQ investment themes, get daily movers, look up which themes a stock belongs to, compare themes side-by-side, and see predictions — all through Claude, Cursor, or any MCP-compatible AI agent.
Powered by TLI (Theme Lifecycle Index) — a Bayesian-optimized scoring algorithm combining search interest, news momentum, market volatility, and stock activity into a 0-100 score with lifecycle stage classification.
11 tools + 4 one-click workflow prompts + live resources. Try the market_briefing slash command for an instant Korean market overview.
Quick Start
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
Cursor / Windsurf
Add to .cursor/mcp.json or .windsurf/mcp.json:
VS Code / Claude Code
Add to .vscode/mcp.json:
Try Asking
After setup, just ask in natural language:
Available Tools
get_theme_ranking
Get theme rankings by lifecycle stage with limit and sort options.
get_market_summary
Get an AI-optimized market overview with top themes (includes themeId for chaining).
get_theme_changes
Get daily or weekly score movers, stage transitions, and newly emerging themes.
compare_themes
Compare 2-5 themes side-by-side with scores, stocks, sparklines, and similarity.
get_predictions
Get themes predicted to rise, peak, or cool based on historical analog matching.
search_themes
Search themes by keyword, stock name, or stock code.
search_stocks
Search stocks by company name or 6-digit code, with related theme preview. Automatically performs stock-to-theme lookup for 6-digit codes.
get_stock_themes
Reverse lookup — every theme a stock belongs to, ranked by theme score, each with lifecycle stage and the stock's relevance. Answers "is my stock in a hot theme?"
get_theme_detail
Get detailed analysis: score breakdown (4 components), stage, prediction, stocks, news, comparisons.
get_theme_history
Get 30-day score history for trend analysis.
get_methodology
Get TLI algorithm documentation — scoring, stages, stabilization, comparison, prediction, data sources, and more.
Tip: Use
section=scoringto get just the scoring algorithm and save context tokens.
Workflow Prompts
One-click slash commands (in Claude Desktop, Cursor, and other MCP clients) that run complete multi-tool analyses for you — the fastest way to get real value:
Resources
Attachable read-only context (no tool call needed):
Scoring Algorithm
TLI scores (0-100) are a weighted sum of 4 components, optimized via Bayesian Optimization:
Scores are stabilized through Cautious Decay (3-signal majority vote), Bollinger Band Clamp (limits daily change), and Age-adaptive EMA (newer themes react faster).
Lifecycle Stages
Stage transitions require 2 consecutive days of the same candidate (hysteresis) and follow Markov transition constraints.
Data Coverage
- 200+ themes across KOSPI & KOSDAQ
- Daily updates — scores, news, stock mappings
- Stock ↔ theme lookup — by company name, 6-digit code, or reverse (stock → all its themes)
- AI market summary for first-call overview
- Predictions with historical analog matching
- Workflow prompts — one-click multi-tool analyses
- Sources: Naver DataLab, Naver Finance, Naver News
Configuration
License
MIT
Links
來源:mcp/README.md,提交 8eb2c73
工具
0版本歷史
1- v0.5.0最新Sep 16, 2026
評分
已滿足適用檢查項的 40%
該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。
- 近 30 天內可連通25 分
請保持端點能夠回應 MCP 握手;SourceWeft 每 30 天探測一次。
- 已宣告認證方式10 分
請在註冊中心條目中宣告憑證請求標頭,或保持端點能夠回應探測,以便判定其認證方式。
- 已列出工具15 分
請讓伺服器在無憑證時回應 tools/list,或發布可在沙箱中無憑證啟動的軟體包。
- 描述必需15 分
- 圖示10 分
- 文件15 分
- 授權條款10 分
請在儲存庫或註冊中心條目中宣告授權條款。