ragalgo-mcp-server 1.0.4 → 1.0.5

package/README.md

@@ -1,73 +1,119 @@
-# RagAlgo: Dynamic RAG Engine for AI Reliability
-
-> **AI 신뢰성을 위한 다이내믹 RAG 엔진**
-
-RagAlgo는 변동성이 큰 한국 금융(주식/코인) 데이터를 **'태그(Tag)'**와 **'점수(Score)'**로 표준화하여 제공하는 **AI 전용 MCP 서버**입니다.
-여러분의 AI 에이전트가 인터넷을 헤매지 않고, 가장 효율적으로 시장의 "진실(State-of-Truth)"을 파악할 수 있도록 돕습니다.
-
-- **핵심 기능:** KOSPI/KOSDAQ 및 업비트(Upbit) 기반의 실시간 차트 점수, 뉴스 감정 분석
-- **무료 테스트:** 지금 바로 설치하여 **1,000회 무료 호출**로 충분히 테스트해보세요.
-- **상세 문서:** 웹소켓 연동 및 비즈니스 플랜 등 자세한 내용은 [공식 웹사이트(ragalgo.com)](https://www.ragalgo.com)에서 확인하실 수 있습니다.
-
----
-
-## 🚀 설치 및 실행
-
-### 1. 직접 실행 (npx)
-
-```bash
-# 별도 설치 없이 바로 실행 (API Key 필요)
-npx -y @ragalgo/server
-```
-
-### 2. Claude Desktop 설정
-
-`%APPDATA%\Claude\claude_desktop_config.json` (Windows) 또는 `~/Library/Application Support/Claude/claude_desktop_config.json` (Mac) 파일에 추가하세요.
-
-```json
-{
-  "mcpServers": {
-    "ragalgo": {
-      "command": "npx",
-      "args": ["-y", "@ragalgo/server", "--stdio"],
-      "env": {
-        "RAGALGO_API_KEY": "비즈니스_또는_테스트_키_입력"
-      }
-    }
-  }
-}
-```
-
-> **Tip:** 테스트용 API Key는 [RagAlgo 대시보드](https://www.ragalgo.com/dashboard)에서 즉시 발급 가능합니다.
-
----
-
-
-
-## 🛠️ 제공 도구 (Tools)
-
-| 도구 | 설명 |
-|------|------|
-| `get_news_scored` | **[핵심]** 감정 점수(Sentiment Score)가 포함된 금융 뉴스 조회 |
-| `get_chart_stock` | **[핵심]** 한국 주식(KOSPI/KOSDAQ) 기술적 분석 점수 조회 |
-| `get_chart_coin` | **[핵심]** 한국 가상화폐(Upbit) 기술적 분석 점수 조회 |
-| `get_snapshots` | 시장 전체 요약 스냅샷 (뉴스 + 차트 + 트렌드) |
-| `get_financials` | 기업 재무제표 (분기/연간 실적) |
-| `search_tags` | 텍스트(종목명)를 RagAlgo 고유 태그로 변환 |
-
----
-
-## 📡 Real-time WebSocket (Business Only)
-
-실시간 주가/코인 점수 데이터를 웹소켓으로 수신할 수 있습니다.
-
-- **대상:** Business Plan 사용자 (월 30개 연결 포함)
-- **주소:** `wss://ragalgo-relay-server-1-production.up.railway.app`
-- **구현 방법:** `socket.io-client` 라이브러리 사용 (자세한 코드는 [공식 문서](https://www.ragalgo.com/docs) 참조)
-
----
-
-## 💬 Community & Support
-
-- **Website:** [ragalgo.com](https://www.ragalgo.com)
-- **Email:** support@ragalgo.com
+# RagAlgo: Dynamic RAG Engine for AI Reliability
+
+> **Global Beta**: Currently specialized in **Korean Finance (Stocks & Crypto)**, but built on a universal "Tag & Score" protocol applicable to any domain.
+
+RagAlgo is not just a data API. It is a **"State-of-Truth" Provider** designed to prevent LLM hallucinations when dealing with **complex static knowledge** and **highly volatile dynamic data**.
+
+### Why RagAlgo?
+Traditional RAG (SQL/Vector DB) often fails to capture the "current context" of fast-changing reality. RagAlgo solves this by converting data into **Mathematical Scores** and **Standardized Tags**.
+
+### Core Philosophy
+1.  **Sanitized Data:** We filter out noise so AI only ingests high-quality signals.
+2.  **Mathematical Scoring:** We translate complex movements into simple scores (e.g., `Chart Score: 9.5`), giving AI a clear ground truth to reason upon.
+3.  **Universal Protocol:** Our "Tag & Score" architecture is designed to **systematize and standardize any data domain**—including Legal, Medical, and Tech—into a format that AI can flawlessly understand and judge.
+
+
+### Flexible Integration: Scored vs. Non-Scored
+We respect your freedom to build. Choose the mode that fits your pipeline:
+
+| Mode | Target User | Description |
+|------|-------------|-------------|
+| **Scored API** (AI-Ready) | Builders & Startups | RagAlgo's math engines sanitize and score everything. Zero hallucinations. |
+| **Non-Scored API** (Raw Mode) | Data Scientists & Engineers | Pure, sanitized data feeds for your custom pipelines. Perfect for fine-tuning, **n8n** automation, **LangChain** agents, or **Flowise** workflows. |
+
+---
+
+# RagAlgo MCP Server (KOR)
+
+금융 뉴스 및 데이터 API를 MCP(Model Context Protocol)로 제공하는 서버입니다.
+
+## 설치
+
+```bash
+cd mcp-server
+npm install
+npm run build
+```
+
+## 환경변수 설정
+
+```bash
+# Windows (PowerShell)
+$env:RAGALGO_API_KEY="ragalgo_scored_test_a3ed9bd570436d46"
+
+# Windows (CMD)
+set RAGALGO_API_KEY=ragalgo_scored_test_a3ed9bd570436d46
+
+# Mac/Linux
+export RAGALGO_API_KEY="ragalgo_scored_test_a3ed9bd570436d46"
+```
+
+## Claude Desktop 설정
+
+`%APPDATA%\Claude\claude_desktop_config.json` 파일에 추가:
+
+```json
+{
+  "mcpServers": {
+    "ragalgo": {
+      "command": "node",
+      "args": ["C:\\coding\\RagAlgo\\mcp-server\\dist\\index.js"],
+      "env": {
+        "RAGALGO_API_KEY": "ragalgo_scored_test_a3ed9bd570436d46"
+      }
+    }
+  }
+}
+```
+
+## 제공 도구 (9개)
+
+| 도구 | 설명 |
+|------|------|
+| `get_news` | 금융 뉴스 조회 (점수 제외) |
+| `get_news_scored` | 금융 뉴스 조회 (점수 포함) |
+| `get_chart_stock` | 주식 차트 점수 조회 |
+| `get_chart_coin` | 코인 차트 점수 조회 |
+| `get_financials` | 재무제표 조회 |
+| `get_snapshots` | 일별 스냅샷 조회 |
+| `search_tags` | 태그 검색 |
+| `match_tags` | 텍스트→태그 매칭 |
+| `get_trends` | 태그별 트렌드 조회 |
+
+## 사용 예시
+
+### Claude Desktop에서
+
+```
+"삼성전자 최근 뉴스 5개 보여줘"
+→ get_news_scored(tag="삼성전자", limit=5)
+
+"반도체 관련 호재 뉴스만 찾아줘"
+→ get_news_scored(tag="반도체", verdict="bullish")
+
+"005930 재무제표 보여줘"
+→ get_financials(ticker="005930")
+
+"비트코인 차트 분석"
+→ get_chart_coin(ticker="KRW-BTC")
+```
+
+## API 키 종류
+
+| Tier | API Key | 접근 가능 도구 |
+|------|---------|---------------|
+| nonscored | `ragalgo_nonscored_test_...` | get_news, get_financials, get_chart_*, search_tags, match_tags, get_trends |
+| scored | `ragalgo_scored_test_...` | 위 전부 + get_news_scored, get_snapshots |
+
+## 개발
+
+```bash
+# 개발 모드 실행
+npm run dev
+
+# 빌드
+npm run build
+
+# 프로덕션 실행
+npm start
+```

package/dist/index.js

@@ -11,6 +11,7 @@ import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js'
 import { SSEServerTransport } from '@modelcontextprotocol/sdk/server/sse.js';
 import express from 'express';
 import cors from 'cors';
+import { zodToJsonSchema } from 'zod-to-json-schema';
 import { CallToolRequestSchema, ListToolsRequestSchema, } from '@modelcontextprotocol/sdk/types.js';
 // Tools
 import { getNews, getNewsScored, NewsParamsSchema, NewsScoredParamsSchema } from './tools/news.js';
@@ -40,25 +41,8 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
                 name: 'search_tags',
                 description: `🔍 [TAG LOOKUP - USE FIRST] ALWAYS use this BEFORE other RagAlgo tools when user mentions any Korean stock, coin, or theme by NAME.
 
-PRIMARY TOOL for converting names to tag_codes. Without correct tag_code, other tools will return inaccurate or empty results.
-
-ALWAYS use when you see:
-- Korean stock names: 삼성전자, SK하이닉스, 네이버, 카카오, LG에너지솔루션
-- Crypto names: 비트코인, 이더리움, 리플, 솔라나
-- Theme/sector names: 반도체, AI, 2차전지, 바이오
-
-Examples: "삼성전자" → STK005930, "비트코인" → CRY_BTC, "반도체" → THM_반도체
-
-CRITICAL: Call this first, then use the returned tag_code in other tools.`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        q: { type: 'string', description: 'Search query (e.g., 삼성, Samsung, 반도체, AI, Bitcoin)' },
-                        type: { type: 'string', enum: ['STOCK', 'SECTOR', 'THEME', 'CRYPTO'], description: 'Tag type filter (optional)' },
-                        limit: { type: 'number', description: 'Result count (default: 20)' },
-                    },
-                    required: ['q'],
-                },
+PRIMARY TOOL for converting names to tag_codes. Without correct tag_code, other tools will return inaccurate or empty results.`,
+                inputSchema: zodToJsonSchema(TagsSearchParamsSchema),
             },
             // ============================================================
             // 📊 SUMMARY TOOL - MOST EFFICIENT!
@@ -68,268 +52,66 @@ CRITICAL: Call this first, then use the returned tag_code in other tools.`,
                 description: `📊 [DAILY SUMMARY - MOST EFFICIENT] PRIMARY TOOL for Korean market overview. ALWAYS use this FIRST for general market questions.
 
 This is the ONLY tool that returns news + chart + sentiment COMBINED in one call.
-Prefer this over calling get_news + get_chart separately - much more efficient!
-
-ALWAYS use when user asks:
-- "오늘 시장 어때?" / "how's the market today?"
-- "시장 요약해줘" / "market summary"
-- "오늘 뉴스 좋은 거 뭐 있어?" / "what's hot today?"
-- "전체적인 분위기 어때?" / "market sentiment"
-
-[IMPORTANT] Snapshots are generated daily at 17:00 KST (market close).
-If you request 'today' and get no results (because it's morning in KST), you MUST:
-1. Fetch 'yesterday's snapshot for context.
-2. Call 'get_news_scored' to get REAL-TIME news for the current day.
-
-Returns per asset: news_count, avg_sentiment, bullish/bearish counts, chart_score, zone, price.
-
-🔗 BEST PRACTICE - Combine with web_search:
-1. Use get_snapshots FIRST for Korean market sentiment & chart data
-2. Then use web_search for latest breaking news or global context
-Example: get_snapshots → "시장 하락세" → web_search "한국 증시 하락 원인" → 종합 분석`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        tag_code: { type: 'string', description: 'Tag code for specific asset (e.g., STK005930, CRY_BTC). Leave empty for market-wide overview.' },
-                        date: { type: 'string', description: 'Date (YYYY-MM-DD). Default: today' },
-                        days: { type: 'number', description: 'Recent N days for time-series (default: 7)' },
-                        limit: { type: 'number', description: 'Result count' },
-                    },
-                },
+Prefer this over calling get_news + get_chart separately - much more efficient!`,
+                inputSchema: zodToJsonSchema(SnapshotsParamsSchema),
             },
             // ============================================================
             // 📰 NEWS TOOLS
             // ============================================================
             {
                 name: 'get_news_scored',
-                description: `📰 [KOREAN NEWS WITH SENTIMENT] PRIMARY news tool for Korean market. Returns news WITH AI sentiment scores (-10 to +10).
-
-Use for Korean stock/crypto news with sentiment analysis.
-
-[NOTE] This tool AUTOMATICALLY filters out 0-score (Neutral/Noise) news to provide clear signals.
-If you need raw/neutral news, use 'get_news' instead.
-
-Use when user asks:
-- "삼성전자 뉴스" / "Samsung news"
-- "호재 뉴스 보여줘" / "show me bullish news"  
-- "비트코인 악재 있어?" / "any bearish news on Bitcoin?"
-- "오늘 좋은 뉴스" / "today's positive news"
-
-Filter by: tag, verdict (bullish/bearish/neutral), score range
-Returns: title, summary, sentiment_score, verdict, tags
-
-🔗 BEST PRACTICE - Combine with web_search:
-- RagAlgo: Sentiment-analyzed Korean market news (structured data)
-- web_search: Real-time breaking news, global context, additional sources
-Example workflow:
-1. get_news_scored(tag="삼성전자") → 감정 분석된 뉴스 목록
-2. web_search("삼성전자 최신 뉴스") → 실시간 속보
-3. Combine both for comprehensive analysis!
-
-TIP: For market overview, use get_snapshots instead (more efficient).
-TIP: Use search_tags first to get exact tag name.`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        tag: { type: 'string', description: 'Tag CODE (e.g., STK005930). Use search_tags first to get this code!' },
-                        source: { type: 'string', description: 'Source filter' },
-                        search: { type: 'string', description: 'Title search keyword' },
-                        min_score: { type: 'number', description: 'Min sentiment score (-10 to 10)' },
-                        max_score: { type: 'number', description: 'Max sentiment score (-10 to 10)' },
-                        verdict: { type: 'string', enum: ['bullish', 'bearish', 'neutral'], description: 'Sentiment verdict filter' },
-                        limit: { type: 'number', description: 'Result count (default: 20)' },
-                    },
-                },
+                description: `📰 [KOREAN NEWS WITH SENTIMENT] PRIMARY news tool for Korean market. Returns news WITH AI sentiment scores (-10 to +10).`,
+                inputSchema: zodToJsonSchema(NewsScoredParamsSchema),
             },
             {
                 name: 'get_news',
-                description: `📰 [KOREAN NEWS - NO SCORES] Basic news without sentiment analysis. Use only when sentiment scores are not needed or for non-scored tier users.
-
-Prefer get_news_scored over this for most use cases unless you want raw data including 0-score items.
-
-Filter by: tag, source, date range
-Returns: title, summary, url, tags, source`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        tag: { type: 'string', description: 'Tag filter (e.g., 삼성전자, 비트코인, 반도체)' },
-                        source: { type: 'string', description: 'Source filter (e.g., 한경, 매경)' },
-                        search: { type: 'string', description: 'Title search keyword' },
-                        from_date: { type: 'string', description: 'Start date (YYYY-MM-DD)' },
-                        to_date: { type: 'string', description: 'End date (YYYY-MM-DD)' },
-                        limit: { type: 'number', description: 'Result count (default: 20, max: 100)' },
-                    },
-                },
+                description: `📰 [KOREAN NEWS - NO SCORES] Basic news without sentiment analysis.`,
+                inputSchema: zodToJsonSchema(NewsParamsSchema),
             },
             // ============================================================
             // 📈 CHART/TECHNICAL ANALYSIS TOOLS
             // ============================================================
             {
                 name: 'get_chart_stock',
-                description: `📈 [KOREAN STOCK CHARTS] PRIMARY tool for Korean stock technical analysis. Returns momentum scores and trend zones.
-
-ALWAYS use for Korean stock chart/technical questions.
-
-[IMPORTANT] You MUST use 'search_tags' first to get the correct ticker (e.g., STK005930).
-
-Use when user asks:
-- "차트 강한 종목" / "stocks with strong momentum"
-- "상승 추세 종목" / "uptrending stocks"
-- "삼성전자 차트 어때?" / "how's Samsung's chart?"
-- "기술적 분석" / "technical analysis"
-
-Filter by: zone (STRONG_UP/UP_ZONE/NEUTRAL/DOWN_ZONE/STRONG_DOWN), market (KOSPI/KOSDAQ)
-Returns: ticker, name, zone, oscillator_state, 5-day scores (d0-d4), last_price
-
-🔗 COMBINE with web_search for deeper analysis:
-1. get_chart_stock → "삼성전자 DOWN_ZONE"
-2. web_search "삼성전자 주가 하락 이유" → 하락 원인 파악
-3. Provide comprehensive technical + fundamental analysis!
-
-TIP: Use search_tags first to get ticker from stock name.`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        ticker: { type: 'string', description: 'Stock ticker (e.g., 005930 for Samsung)' },
-                        market: { type: 'string', enum: ['KOSPI', 'KOSDAQ'], description: 'Market type' },
-                        zone: { type: 'string', enum: ['STRONG_UP', 'UP_ZONE', 'NEUTRAL', 'DOWN_ZONE', 'STRONG_DOWN'], description: 'Chart zone filter - use this to find strong/weak stocks' },
-                        limit: { type: 'number', description: 'Result count' },
-                    },
-                },
+                description: `📈 [KOREAN STOCK CHARTS] PRIMARY tool for Korean stock technical analysis. Returns momentum scores and trend zones.`,
+                inputSchema: zodToJsonSchema(ChartStockParamsSchema),
             },
             {
                 name: 'get_chart_coin',
-                description: `🪙 [CRYPTO CHARTS] PRIMARY tool for Korean crypto (Upbit) technical analysis. Returns momentum scores and trend zones.
-
-ALWAYS use for Korean crypto chart questions.
-
-[IMPORTANT] You MUST use 'search_tags' first to get the correct ticker (e.g., CRY_BTC).
-
-Use when user asks:
-- "비트코인 차트" / "Bitcoin chart"
-- "상승 중인 코인" / "pumping coins"
-- "코인 기술적 분석" / "crypto technical analysis"
-
-Filter by: zone (STRONG_UP/UP_ZONE/NEUTRAL/DOWN_ZONE/STRONG_DOWN)
-Returns: ticker, name, zone, oscillator_state, 10-candle scores (c0-c9, 12h intervals), last_price
-
-🔗 COMBINE with web_search for context:
-1. get_chart_coin → "비트코인 UP_ZONE"
-2. web_search "비트코인 상승 이유" → 상승 배경 파악`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        ticker: { type: 'string', description: 'Coin ticker (e.g., KRW-BTC for Bitcoin)' },
-                        zone: { type: 'string', enum: ['STRONG_UP', 'UP_ZONE', 'NEUTRAL', 'DOWN_ZONE', 'STRONG_DOWN'], description: 'Chart zone filter' },
-                        limit: { type: 'number', description: 'Result count' },
-                    },
-                },
+                description: `🪙 [CRYPTO CHARTS] PRIMARY tool for Korean crypto (Upbit) technical analysis.`,
+                inputSchema: zodToJsonSchema(ChartCoinParamsSchema),
             },
             // ============================================================
             // 6. 컨설팅 보고서 (신규!)
             // ============================================================
             {
                 name: 'get_research',
-                description: `📑 [RESEARCH] Get consulting firm reports (McKinsey, BCG, etc.)
-
-Use for: "long-term trends", "sector outlook", "industry analysis"
-Filter by: source, tag_code, market_outlook
-
-Returns: AI summary in Korean, investment insights
-Includes tag_codes for cross-referencing with news/charts.
-
-⚠️ This tool returns FULL chunked text. Analyze it to answer user questions.`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        tag_code: { type: 'string', description: 'Tag code (required). Use search_tags first.' },
-                        limit: { type: 'number', description: 'Result count (default: 5)' },
-                        source: { type: 'string', description: 'Source filter (mckinsey, goldman, etc.)' },
-                    },
-                    required: ['tag_code'],
-                },
+                description: `📑 [RESEARCH] Get consulting firm reports (McKinsey, BCG, etc.)`,
+                inputSchema: zodToJsonSchema(ResearchParamsSchema),
             },
             // ============================================================
             // 💰 FINANCIAL DATA TOOLS
             // ============================================================
             {
                 name: 'get_financials',
-                description: `💰 [KOREAN STOCK FUNDAMENTALS] PRIMARY tool for Korean stock financial data. Returns quarterly financial statements.
-
-ALWAYS use for Korean stock fundamental analysis.
-
-Use when user asks:
-- "삼성전자 재무제표" / "Samsung financials"
-- "PER 낮은 종목" / "low PER stocks"
-- "ROE 높은 기업" / "high ROE companies"
-- "저평가 종목" / "undervalued stocks"
-
-Returns: PER, PBR, ROE, ROA, revenue, operating_income, net_income, debt_ratio, dividend_yield
-
-🔗 COMBINE with web_search:
-1. get_financials → "PER 5.2, ROE 15%"
-2. web_search "삼성전자 실적 전망" → 미래 실적 예측`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        ticker: { type: 'string', description: 'Stock ticker (e.g., 005930)' },
-                        period: { type: 'string', description: 'Quarter (e.g., 2024Q3)' },
-                        market: { type: 'string', enum: ['KOSPI', 'KOSDAQ'], description: 'Market type' },
-                        periods: { type: 'number', description: 'Recent N quarters (default: 4)' },
-                        limit: { type: 'number', description: 'Result count' },
-                    },
-                },
+                description: `💰 [KOREAN STOCK FUNDAMENTALS] PRIMARY tool for Korean stock financial data.`,
+                inputSchema: zodToJsonSchema(FinancialsParamsSchema),
             },
             // ============================================================
             // 📉 TREND TOOLS
             // ============================================================
             {
                 name: 'get_trends',
-                description: `📉 [SENTIMENT TRENDS] Get historical sentiment trend for a specific asset over time.
-
-Use when user asks:
-- "삼성전자 지난주 분위기" / "Samsung sentiment last week"
-- "비트코인 추세" / "Bitcoin trend"
-- "최근 7일간 뉴스 동향" / "news trend over 7 days"
-
-REQUIRES tag_code - use search_tags first!
-Returns: daily news_count and avg_sentiment_score over N days
-
-🔗 COMBINE with web_search:
-1. get_trends → "지난주 감정 -2.5로 하락"
-2. web_search "삼성전자 지난주 이슈" → 하락 원인 파악`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        tag_code: { type: 'string', description: 'Tag code (e.g., STK005930, CRY_BTC) - REQUIRED. Use search_tags to find this first!' },
-                        days: { type: 'number', description: 'Recent N days (default: 7, max: 30)' },
-                    },
-                    required: ['tag_code'],
-                },
+                description: `📉 [SENTIMENT TRENDS] Get historical sentiment trend for a specific asset over time.`,
+                inputSchema: zodToJsonSchema(TrendsParamsSchema),
             },
             // ============================================================
             // 🏷️ AUTO-TAGGING TOOL
             // ============================================================
             {
                 name: 'match_tags',
-                description: `🏷️ [AUTO-TAG EXTRACTION] Extract stock/crypto/theme tags from any text. Useful for categorizing news or analyzing what topics a text mentions.
-
-Use when:
-- Analyzing what stocks/themes a news title mentions
-- Auto-categorizing text content
-- Finding related tags from a sentence
-
-Input: any text (e.g., "삼성전자 HBM 대박 소식")
-Returns: matched tags with confidence scores`,
-                inputSchema: {
-                    type: 'object',
-                    properties: {
-                        text: { type: 'string', description: 'Text to analyze (e.g., "삼성전자 HBM 대박 소식")' },
-                        types: { type: 'array', items: { type: 'string' }, description: 'Tag type filter (optional)' },
-                        limit: { type: 'number', description: 'Result count (default: 10)' },
-                    },
-                    required: ['text'],
-                },
+                description: `🏷️ [AUTO-TAG EXTRACTION] Extract stock/crypto/theme tags from any text.`,
+                inputSchema: zodToJsonSchema(TagsMatchParamsSchema),
             },
         ],
     };
@@ -396,6 +178,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
     }
 });
 // Start server
+// Start server
 async function main() {
     // Check if running in stdio mode (command line argument or specific env var)
     const isStdio = process.argv.includes('--stdio');

package/dist/tools/chart.d.ts

@@ -4,33 +4,30 @@
 import { z } from 'zod';
 export declare const ChartStockParamsSchema: z.ZodObject<{
     ticker: z.ZodOptional<z.ZodString>;
-    market: z.ZodOptional<z.ZodEnum<["KOSPI", "KOSDAQ"]>>;
-    zone: z.ZodOptional<z.ZodEnum<["STRONG_UP", "UP_ZONE", "NEUTRAL", "DOWN_ZONE", "STRONG_DOWN"]>>;
+    market: z.ZodOptional<z.ZodEnum<{
+        KOSPI: "KOSPI";
+        KOSDAQ: "KOSDAQ";
+    }>>;
+    zone: z.ZodOptional<z.ZodEnum<{
+        STRONG_UP: "STRONG_UP";
+        UP_ZONE: "UP_ZONE";
+        NEUTRAL: "NEUTRAL";
+        DOWN_ZONE: "DOWN_ZONE";
+        STRONG_DOWN: "STRONG_DOWN";
+    }>>;
     limit: z.ZodDefault<z.ZodNumber>;
-}, "strip", z.ZodTypeAny, {
-    limit: number;
-    ticker?: string | undefined;
-    market?: "KOSPI" | "KOSDAQ" | undefined;
-    zone?: "STRONG_UP" | "UP_ZONE" | "NEUTRAL" | "DOWN_ZONE" | "STRONG_DOWN" | undefined;
-}, {
-    limit?: number | undefined;
-    ticker?: string | undefined;
-    market?: "KOSPI" | "KOSDAQ" | undefined;
-    zone?: "STRONG_UP" | "UP_ZONE" | "NEUTRAL" | "DOWN_ZONE" | "STRONG_DOWN" | undefined;
-}>;
+}, z.core.$strip>;
 export declare const ChartCoinParamsSchema: z.ZodObject<{
     ticker: z.ZodOptional<z.ZodString>;
-    zone: z.ZodOptional<z.ZodEnum<["STRONG_UP", "UP_ZONE", "NEUTRAL", "DOWN_ZONE", "STRONG_DOWN"]>>;
+    zone: z.ZodOptional<z.ZodEnum<{
+        STRONG_UP: "STRONG_UP";
+        UP_ZONE: "UP_ZONE";
+        NEUTRAL: "NEUTRAL";
+        DOWN_ZONE: "DOWN_ZONE";
+        STRONG_DOWN: "STRONG_DOWN";
+    }>>;
     limit: z.ZodDefault<z.ZodNumber>;
-}, "strip", z.ZodTypeAny, {
-    limit: number;
-    ticker?: string | undefined;
-    zone?: "STRONG_UP" | "UP_ZONE" | "NEUTRAL" | "DOWN_ZONE" | "STRONG_DOWN" | undefined;
-}, {
-    limit?: number | undefined;
-    ticker?: string | undefined;
-    zone?: "STRONG_UP" | "UP_ZONE" | "NEUTRAL" | "DOWN_ZONE" | "STRONG_DOWN" | undefined;
-}>;
+}, z.core.$strip>;
 export type ChartStockParams = z.infer<typeof ChartStockParamsSchema>;
 export type ChartCoinParams = z.infer<typeof ChartCoinParamsSchema>;
 export declare function getChartStock(params: ChartStockParams): Promise<{
@@ -42,12 +39,6 @@ export declare function getChartStock(params: ChartStockParams): Promise<{
         scores: number[];
         zone: string;
         oscillator_state: string;
-        analysis?: {
-            weekly_zone: string;
-            daily_momentum: string;
-            daily_signal: string;
-        };
-        outlook?: string;
         last_price: number;
         updated_at: string;
     } | Array<{
@@ -66,12 +57,6 @@ export declare function getChartCoin(params: ChartCoinParams): Promise<{
         scores: number[];
         zone: string;
         oscillator_state: string;
-        analysis?: {
-            weekly_zone: string;
-            daily_momentum: string;
-            daily_signal: string;
-        };
-        outlook?: string;
         last_price: number;
         updated_at: string;
     } | Array<{

package/dist/tools/financials.d.ts

@@ -5,25 +5,14 @@ import { z } from 'zod';
 export declare const FinancialsParamsSchema: z.ZodObject<{
     ticker: z.ZodOptional<z.ZodString>;
     period: z.ZodOptional<z.ZodString>;
-    market: z.ZodOptional<z.ZodEnum<["KOSPI", "KOSDAQ"]>>;
+    market: z.ZodOptional<z.ZodEnum<{
+        KOSPI: "KOSPI";
+        KOSDAQ: "KOSDAQ";
+    }>>;
     periods: z.ZodDefault<z.ZodNumber>;
     limit: z.ZodDefault<z.ZodNumber>;
     offset: z.ZodDefault<z.ZodNumber>;
-}, "strip", z.ZodTypeAny, {
-    limit: number;
-    offset: number;
-    periods: number;
-    ticker?: string | undefined;
-    market?: "KOSPI" | "KOSDAQ" | undefined;
-    period?: string | undefined;
-}, {
-    limit?: number | undefined;
-    offset?: number | undefined;
-    ticker?: string | undefined;
-    market?: "KOSPI" | "KOSDAQ" | undefined;
-    period?: string | undefined;
-    periods?: number | undefined;
-}>;
+}, z.core.$strip>;
 export type FinancialsParams = z.infer<typeof FinancialsParamsSchema>;
 export declare function getFinancials(params: FinancialsParams): Promise<{
     success: boolean;