edition-mcp-server 0.1.4 → 0.2.0

package/README.md

@@ -1,15 +1,11 @@
-# edition-mcp-server
+# EDITION Intelligence Platform — MCP Server
 
-MCP server for the **EDITION Intelligence Platform** — contextual intelligence infrastructure for AI agents operating in Japan.
+Japan Operations OS for autonomous AI agents. 14 knowledge domains, 31 MCP tools, 2 prompt templates — verified, structured intelligence for agents operating in Japan.
 
-## What it does
+[![npm](https://img.shields.io/npm/v/edition-mcp-server)](https://www.npmjs.com/package/edition-mcp-server)
+[![License: MIT](https://img.shields.io/badge/License-MIT-black.svg)](https://opensource.org/licenses/MIT)
 
-Provides 8 tools that give AI agents the ability to:
-
-- **Remember** conversations with Japanese-aware fact extraction (keigo analysis, social hierarchy detection, confidence scoring)
-- **Navigate** Japanese business regulations across 10 industries + tourist rules
-
-## Install
+## Installation
 
 ```bash
 npx edition-mcp-server
@@ -22,8 +18,6 @@ npm install -g edition-mcp-server
 edition-mcp
 ```
 
-## Configure
-
 ### Claude Desktop
 
 Add to `claude_desktop_config.json`:
@@ -33,19 +27,28 @@ Add to `claude_desktop_config.json`:
   "mcpServers": {
     "edition": {
       "command": "npx",
-      "args": ["-y", "edition-mcp-server"],
-      "env": {
-        "EDITION_API_URL": "http://localhost:8000",
-        "EDITION_API_KEY": "your_api_key"
-      }
+      "args": ["-y", "edition-mcp-server"]
     }
   }
 }
 ```
 
-### Cursor
+### Cursor / Windsurf / VS Code
 
-Add to MCP settings:
+Add to your MCP settings:
+
+```json
+{
+  "edition": {
+    "command": "npx",
+    "args": ["-y", "edition-mcp-server"]
+  }
+}
+```
+
+### Custom API URL
+
+By default connects to `https://api.edition.sh`. Override with environment variables:
 
 ```json
 {
@@ -53,8 +56,8 @@ Add to MCP settings:
     "command": "npx",
     "args": ["-y", "edition-mcp-server"],
     "env": {
-      "EDITION_API_URL": "http://localhost:8000",
-      "EDITION_API_KEY": "your_api_key"
+      "EDITION_API_URL": "https://api.edition.sh",
+      "EDITION_API_KEY": "edition_dev_key_for_testing"
     }
   }
 }
@@ -62,28 +65,134 @@ Add to MCP settings:
 
 ## Tools
 
+### Memory (5 tools)
+
 | Tool | Description |
 |------|-------------|
-| `memory_store` | Store episode + auto-extract structured facts from Japanese text |
-| `memory_recall` | Semantic search across stored episodes |
-| `memory_facts` | List structured facts (subject → predicate → object with confidence) |
-| `memory_context` | Get context summary for an entity or session |
-| `memory_extract` | Extract facts from text with keigo/hierarchy analysis |
-| `regulation_check` | Check Japanese business regulations by industry |
-| `regulation_industries` | List all 10 supported industries |
-| `regulation_tourist` | Tourist regulation lookup (visa, tax-free, transit, etc.) |
+| `memory_store` | Store episodes to persistent memory with Japanese language understanding (keigo, subject omission, implicit context) |
+| `memory_recall` | Semantic search across stored memories — handles ambiguous Japanese queries |
+| `memory_facts` | List structured facts as subject → predicate → object triples with confidence scores |
+| `memory_context` | Get current session context summary for prompt injection |
+| `memory_extract` | Auto-extract structured facts from text with keigo/hierarchy analysis |
 
-## Backend
+### Regulations (3 tools)
 
-This MCP server connects to the EDITION REST API backend. To run your own:
+| Tool | Description |
+|------|-------------|
+| `regulation_check` | Check Japan business regulations — licenses, governing bodies, penalties, costs, timelines for 10 industries |
+| `regulation_industries` | List all 10 regulated industries with their requirements |
+| `regulation_tourist` | Tourist/visitor regulation categories (visa, tax-free, transport, accommodation, medical, manners) |
 
-```bash
-git clone https://github.com/hiroshic9-png/edition.git
-cd edition
-python3 -m venv venv && source venv/bin/activate
-pip install -r requirements.txt
-python -m uvicorn backend.api.main:app --reload
-```
+### Protocols (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `protocol_check` | Search Japanese business protocols — nemawashi, ringi, hourensou, meishi exchange, seating order, gift-giving |
+| `protocol_list` | List all business protocols with importance levels |
+
+### Calendar (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `calendar_check` | Search Japan business calendar — fiscal year, holidays, gift seasons, deadlines, seasonal patterns |
+| `calendar_list` | List all calendar categories |
+
+### Regional (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `regional_check` | Search regional business differences — city characteristics, subsidies, local regulations, business customs |
+| `regional_list` | List all regional categories |
+
+### Organization (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `organization_check` | Search organizational structures — keiretsu, corporate hierarchy, payment customs, contract practices |
+| `organization_list` | List all organization categories |
+
+### Foreign Entry (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `foreign_entry_check` | Japan market entry guides — incorporation, management visa, bank account, real estate, tax registration |
+| `foreign_entry_list` | List all foreign entry categories |
+
+### Travel (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `travel_search` | Search Japan travel knowledge — shinkansen, IC cards, ryokan etiquette, tipping culture |
+| `travel_list` | List all travel topics |
+
+### Entertainment (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `entertainment_search` | Search Japan entertainment — oshi-katsu fan culture, anime pilgrimage, live events, seasonal festivals |
+| `entertainment_list` | List all entertainment topics |
+
+### Daily Life (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `daily_life_search` | Search daily life in Japan — postal/address system, garbage sorting, utilities, healthcare/insurance |
+| `daily_life_list` | List all daily life topics |
+
+### Language (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `language_search` | Search Japanese language structures — keigo system, counter words (josushi), name patterns, business templates |
+| `language_list` | List all language topics |
+
+### Food Culture (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `food_search` | Search Japan food culture — dining etiquette, cuisine types, restaurant navigation, dietary restrictions (halal, vegetarian) |
+| `food_list` | List all food culture topics |
+
+### Disaster & Safety (2 tools)
+
+| Tool | Description |
+|------|-------------|
+| `disaster_search` | Search disaster/safety knowledge — earthquake shindo scale, typhoon warnings, emergency contacts (110/119), disaster preparedness |
+| `disaster_list` | List all disaster & safety topics |
+
+### Cross-Domain (1 tool)
+
+| Tool | Description |
+|------|-------------|
+| `search` | Cross-domain search across all 14 knowledge domains simultaneously — one query returns matches from all domains |
+
+## Prompt Templates
+
+| Prompt | Description |
+|--------|-------------|
+| `japan_business_briefing` | Generate a comprehensive Japan business briefing for a specific industry |
+| `japan_travel_guide` | Generate a practical travel guide for a specific Japan destination |
+
+## Resources
+
+| Resource | Description |
+|----------|-------------|
+| `edition://domains` | List of all 14 knowledge domains with descriptions and endpoint information |
+| `edition://quality` | Trust Anchor quality scores for each domain — verified data coverage metrics |
+
+## Architecture
+
+EDITION uses a three-layer knowledge architecture:
+
+1. **Rules** — Official regulations, procedures, and requirements
+2. **Context** — Unwritten rules, cultural "why", and implicit expectations
+3. **Experience** — How to turn correct execution into relationship-building moments
+
+## API
+
+- **REST API**: [https://api.edition.sh](https://api.edition.sh)
+- **API Docs**: [https://api.edition.sh/docs](https://api.edition.sh/docs)
+- **Landing Page**: [https://edition.sh](https://edition.sh)
 
 ## License
 

package/dist/index.js

@@ -47,7 +47,7 @@ async function apiPost(path, body) {
 // ── MCP Server ──────────────────────────────────────
 const server = new mcp_js_1.McpServer({
     name: "edition",
-    version: "0.1.2",
+    version: "0.2.0",
 });
 // ── Tool: memory_store ──────────────────────────────
 server.tool("memory_store", "会話やイベントのエピソードを永続記憶に保存します。日本語の文脈（敬語レベル、主語省略、暗黙の了解）も構造化して保持します。auto_extract=trueにすると、テキストからファクト（主語→述語→目的語の三つ組）を自動抽出します。", {
@@ -380,9 +380,101 @@ server.tool("entertainment_list", "日本のエンタメ知識のトピック一
     }
     return { content: [{ type: "text", text }] };
 });
+// ── Tool: daily_life_search ─────────────────────────
+server.tool("daily_life_search", "日本の日常生活に関する知識を検索します。住所・郵便システム、ゴミ分別ルール、公共料金（電気・ガス・水道・NHK）、医療・健康保険制度。外国人が日本で生活するために必要な実用知識。", {
+    query: zod_1.z.string().describe("検索クエリ（例: 'ゴミの分別方法', '健康保険の加入', '引っ越しの手続き'）"),
+}, async ({ query }) => {
+    const result = await apiPost("/api/v1/daily-life/search", { query });
+    if (!result.results?.length) {
+        return { content: [{ type: "text", text: `❌ '${query}' に該当する日常生活情報が見つかりませんでした。` }] };
+    }
+    let text = `🏠 日常生活情報 (${result.total_matches}件ヒット):\n\n`;
+    for (const r of result.results) {
+        text += `  📌 ${r.name_ja} (${r.name_en})\n     ${r.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: daily_life_list ──────────────────────────
+server.tool("daily_life_list", "日本の日常生活知識のトピック一覧を取得します。", {}, async () => {
+    const result = await apiGet("/api/v1/daily-life/list");
+    let text = `🏠 日常生活トピック一覧 (${result.total}件):\n\n`;
+    for (const t of result.topics) {
+        text += `  • ${t.name_ja} (${t.name_en})\n    ${t.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: language_search ──────────────────────────
+server.tool("language_search", "日本語の構造的知識を検索します。敬語体系（尊敬語・謙譲語・丁寧語）、助数詞（数え方）、名前・住所の構造パターン、ビジネス日本語（電話応対・クッション言葉・メールテンプレート）。", {
+    query: zod_1.z.string().describe("検索クエリ（例: '敬語の使い方', '助数詞の一覧', 'ビジネスメールの書き方'）"),
+}, async ({ query }) => {
+    const result = await apiPost("/api/v1/language/search", { query });
+    if (!result.results?.length) {
+        return { content: [{ type: "text", text: `❌ '${query}' に該当する日本語情報が見つかりませんでした。` }] };
+    }
+    let text = `🗾 日本語知識 (${result.total_matches}件ヒット):\n\n`;
+    for (const r of result.results) {
+        text += `  📌 ${r.name_ja} (${r.name_en})\n     ${r.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: language_list ────────────────────────────
+server.tool("language_list", "日本語知識のトピック一覧を取得します。", {}, async () => {
+    const result = await apiGet("/api/v1/language/list");
+    let text = `🗾 日本語トピック一覧 (${result.total}件):\n\n`;
+    for (const t of result.topics) {
+        text += `  • ${t.name_ja} (${t.name_en})\n    ${t.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: food_search ──────────────────────────────
+server.tool("food_search", "日本の食文化に関する知識を検索します。食事マナー（箸のタブー・乾杯・割り勘）、料理分類（懐石・定食・ラーメン・郷土料理）、飲食店ガイド（食券機・居酒屋・回転寿司・おまかせ）、アレルギー・食制限（ハラル・ベジタリアン対応）。", {
+    query: zod_1.z.string().describe("検索クエリ（例: '箸のマナー', 'ハラル対応レストラン', '回転寿司の注文方法'）"),
+}, async ({ query }) => {
+    const result = await apiPost("/api/v1/food/search", { query });
+    if (!result.results?.length) {
+        return { content: [{ type: "text", text: `❌ '${query}' に該当する食文化情報が見つかりませんでした。` }] };
+    }
+    let text = `🍣 食文化情報 (${result.total_matches}件ヒット):\n\n`;
+    for (const r of result.results) {
+        text += `  📌 ${r.name_ja} (${r.name_en})\n     ${r.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: food_list ────────────────────────────────
+server.tool("food_list", "日本の食文化知識のトピック一覧を取得します。", {}, async () => {
+    const result = await apiGet("/api/v1/food/list");
+    let text = `🍣 食文化トピック一覧 (${result.total}件):\n\n`;
+    for (const t of result.topics) {
+        text += `  • ${t.name_ja} (${t.name_en})\n    ${t.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: disaster_search ──────────────────────────
+server.tool("disaster_search", "日本の災害・安全に関する知識を検索します。地震（震度スケール・緊急地震速報・耐震基準）、台風・水害（警戒レベル・計画運休）、緊急連絡先（110/119/多言語対応）、防災準備（防災バッグ・ハザードマップ・避難所マナー）。", {
+    query: zod_1.z.string().describe("検索クエリ（例: '地震が来たらどうする', '緊急連絡先', '防災バッグの中身'）"),
+}, async ({ query }) => {
+    const result = await apiPost("/api/v1/disaster/search", { query });
+    if (!result.results?.length) {
+        return { content: [{ type: "text", text: `❌ '${query}' に該当する災害・安全情報が見つかりませんでした。` }] };
+    }
+    let text = `⚠️ 災害・安全情報 (${result.total_matches}件ヒット):\n\n`;
+    for (const r of result.results) {
+        text += `  📌 ${r.name_ja} (${r.name_en})\n     ${r.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
+// ── Tool: disaster_list ────────────────────────────
+server.tool("disaster_list", "日本の災害・安全知識のトピック一覧を取得します。", {}, async () => {
+    const result = await apiGet("/api/v1/disaster/list");
+    let text = `⚠️ 災害・安全トピック一覧 (${result.total}件):\n\n`;
+    for (const t of result.topics) {
+        text += `  • ${t.name_ja} (${t.name_en})\n    ${t.summary}\n\n`;
+    }
+    return { content: [{ type: "text", text }] };
+});
 // ── Tool: search ────────────────────────────────────
-server.tool("search", "EDITION全ドメインを横断検索します。1回のリクエストで規制・プロトコル・カレンダー・地域・組織・進出手続き・旅行・エンタメの全8ドメインを同時検索。", {
-    query: zod_1.z.string().describe("検索クエリ（例: '大阪で飲食店を開業', '推し活のマナー', '外国人のビザ取得'）"),
+server.tool("search", "EDITION全14ドメインを横断検索します。1回のリクエストで規制・プロトコル・カレンダー・地域・組織・進出手続き・旅行・エンタメ・日常生活・日本語・食文化・災害安全の全12ドメインを同時検索。", {
+    query: zod_1.z.string().describe("検索クエリ（例: '大阪で飲食店を開業', '地震の避難方法', '敬語の使い方'）"),
 }, async ({ query }) => {
     const result = await apiPost("/api/v1/search", { query });
     let text = `🔍 横断検索結果: ${result.domains_matched}/${result.domains_searched} ドメインでヒット\n\n`;
@@ -397,6 +489,65 @@ server.tool("search", "EDITION全ドメインを横断検索します。1回の
     }
     return { content: [{ type: "text", text }] };
 });
+// ── Resources ───────────────────────────────────────
+server.resource("domains", "edition://domains", {
+    description: "All 14 knowledge domains with descriptions, endpoints, and coverage status",
+    mimeType: "application/json",
+}, async () => {
+    const domains = [
+        { id: "memory", name: "Persistent Memory", endpoint: "/api/v1/memory", tools: 5, layers: ["rules", "context"] },
+        { id: "regulation", name: "Business Regulations", endpoint: "/api/v1/regulation", tools: 3, layers: ["rules", "context", "experience"] },
+        { id: "protocol", name: "Business Protocols", endpoint: "/api/v1/protocol", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "calendar", name: "Business Calendar", endpoint: "/api/v1/calendar", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "regional", name: "Regional Intelligence", endpoint: "/api/v1/regional", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "organization", name: "Organizational Structures", endpoint: "/api/v1/organization", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "foreign_entry", name: "Foreign Market Entry", endpoint: "/api/v1/foreign-entry", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "travel", name: "Travel Intelligence", endpoint: "/api/v1/travel", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "entertainment", name: "Entertainment & Pop Culture", endpoint: "/api/v1/entertainment", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "daily_life", name: "Daily Life", endpoint: "/api/v1/daily-life", tools: 2, layers: ["rules", "context"] },
+        { id: "language", name: "Japanese Language", endpoint: "/api/v1/language", tools: 2, layers: ["rules", "context"] },
+        { id: "food", name: "Food Culture", endpoint: "/api/v1/food", tools: 2, layers: ["rules", "context", "experience"] },
+        { id: "disaster", name: "Disaster & Safety", endpoint: "/api/v1/disaster", tools: 2, layers: ["rules", "context"] },
+        { id: "search", name: "Cross-Domain Search", endpoint: "/api/v1/search", tools: 1, layers: ["rules", "context", "experience"] },
+    ];
+    return {
+        contents: [
+            {
+                uri: "edition://domains",
+                mimeType: "application/json",
+                text: JSON.stringify({ total: domains.length, total_tools: 31, domains }, null, 2),
+            },
+        ],
+    };
+});
+server.resource("quality", "edition://quality", {
+    description: "Trust Anchor quality scores — verified data coverage, source reliability, and 3-layer completeness for each domain",
+    mimeType: "application/json",
+}, async () => {
+    try {
+        const data = await apiGet("/api/v1/analytics/quality");
+        return {
+            contents: [
+                {
+                    uri: "edition://quality",
+                    mimeType: "application/json",
+                    text: JSON.stringify(data, null, 2),
+                },
+            ],
+        };
+    }
+    catch {
+        return {
+            contents: [
+                {
+                    uri: "edition://quality",
+                    mimeType: "application/json",
+                    text: JSON.stringify({ error: "Quality endpoint unavailable" }),
+                },
+            ],
+        };
+    }
+});
 // ── Start ───────────────────────────────────────────
 async function main() {
     const transport = new stdio_js_1.StdioServerTransport();

package/package.json

@@ -1,8 +1,8 @@
 {
   "name": "edition-mcp-server",
-  "version": "0.1.4",
+  "version": "0.2.0",
   "mcpName": "io.github.hiroshic9-png/japan-business-operations",
-  "description": "EDITION Intelligence Platform \u2014 MCP server for Japan operations intelligence (10 knowledge domains)",
+  "description": "EDITION Intelligence Platform — MCP server for Japan operations intelligence. 14 knowledge domains, 31 tools, 2 prompts, 2 resources. Verified ground truth: regulations, procedures, protocols, calendar, regional, organization, foreign entry, travel, entertainment, daily life, language, food culture, disaster safety, and persistent memory.",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
   "bin": {
@@ -24,8 +24,22 @@
     "travel",
     "entertainment",
     "protocol",
-    "calendar"
+    "calendar",
+    "keigo",
+    "business",
+    "culture",
+    "food",
+    "disaster",
+    "safety",
+    "language",
+    "knowledge-base"
   ],
+  "homepage": "https://edition.sh",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/hiroshic9-png/edition-api"
+  },
+  "author": "EDITION <h.sato@c-9.co.jp>",
   "license": "MIT",
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.12.1"