@kiyeonjeon21/datacontext 0.2.0 → 0.3.1

package/.cursorrules

@@ -4,6 +4,8 @@
 MCP server + REST API + SDK for AI-native database context.
 Open-source, offline-first. Business docs in `../datacontext/`.
 
+**중요:** 코드 변경 시 `docs/SYNC_GUIDE.md`를 참고하여 관련 컴포넌트(VSCode Extension, 문서 등)도 함께 업데이트하세요.
+
 ## Architecture
 ```
 src/
@@ -51,6 +53,16 @@ TypeScript (strict), Node.js 18+, npm, Vitest
 1. Add route in `src/api/server.ts`
 2. Use `service` (DataContextService), not adapters directly
 3. Update `docs/API.md`
+4. **Sync:** Update `vscode-datacontext/src/services/connectionManager.ts`
+5. **Sync:** Update `vscode-datacontext/README.md` API Endpoints 섹션
+
+### Sync Checklist (중요!)
+REST API/MCP Tool 변경 시 아래 파일들도 확인:
+- `vscode-datacontext/src/services/connectionManager.ts` — API 메서드
+- `vscode-datacontext/src/extension.ts` — 명령어 핸들러
+- `vscode-datacontext/package.json` — commands 정의
+- `vscode-datacontext/README.md` — 문서
+- 자세한 내용: `docs/SYNC_GUIDE.md`
 
 ## Testing
 ```bash

package/.env.example

@@ -0,0 +1,8 @@
+# DataContext Environment Variables
+# Copy this file to .env and fill in your values
+
+# Claude API Key for LLM features (required for auto-glossary generation)
+ANTHROPIC_API_KEY=your-api-key-here
+
+# Optional: Default model
+# ANTHROPIC_MODEL=claude-sonnet-4-20250514

package/.github/workflows/ci.yml

@@ -10,6 +10,21 @@ jobs:
   test:
     runs-on: ubuntu-latest
     
+    services:
+      postgres:
+        image: postgres:15
+        env:
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: datacontext_test
+        ports:
+          - 5432:5432
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+    
     strategy:
       matrix:
         node-version: [18.x, 20.x]
@@ -31,6 +46,12 @@ jobs:
       
       - name: Run tests
         run: npm test
+        env:
+          PGHOST: localhost
+          PGPORT: 5432
+          PGUSER: postgres
+          PGPASSWORD: postgres
+          PGDATABASE: datacontext_test
 
   lint:
     runs-on: ubuntu-latest
@@ -49,4 +70,3 @@ jobs:
       
       - name: Build
         run: npm run build
-

package/.github/workflows/publish.yml

@@ -8,6 +8,21 @@ jobs:
   publish:
     runs-on: ubuntu-latest
     
+    services:
+      postgres:
+        image: postgres:15
+        env:
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: datacontext_test
+        ports:
+          - 5432:5432
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+    
     permissions:
       contents: read
       id-token: write
@@ -30,9 +45,14 @@ jobs:
       
       - name: Run tests
         run: npm test
+        env:
+          PGHOST: localhost
+          PGPORT: 5432
+          PGUSER: postgres
+          PGPASSWORD: postgres
+          PGDATABASE: datacontext_test
       
       - name: Publish to npm
         run: npm publish --access public
         env:
           NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
-

package/CHANGELOG.md

@@ -7,6 +7,47 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.0] - 2025-01-01
+
+### Added
+- **Business Glossary System**: Define custom business terms and let AI use them
+  - `glossary add` - Add terms manually with SQL expressions
+  - `glossary list` - List all terms with filtering
+  - `glossary search` - Search for terms by name/synonyms
+  - `glossary generate` - AI-powered term generation from natural language
+  - `glossary export` - Export to YAML/JSON for version control
+  - `glossary import` - Import from YAML/JSON files
+  - `glossary delete` - Remove terms
+- **Query Enhancement**: `enhance_query` MCP tool and REST endpoint
+  - Matches natural language queries to glossary terms
+  - Returns SQL conditions for matched terms
+  - Uses local matching first, then AI if available
+- **REST API Endpoints**:
+  - `GET /api/terms` - List all terms
+  - `GET /api/terms/search?q=` - Search terms
+  - `POST /api/terms` - Add term manually
+  - `POST /api/terms/generate` - AI-powered generation
+  - `POST /api/terms/enhance` - Enhance query with terms
+  - `DELETE /api/terms/:id` - Delete term
+- **MCP Tools**: `generate_glossary`, `add_term`, `list_terms`, `search_terms`, `enhance_query`
+- API server start script: `npm run serve` for standalone server
+- Claude API integration for AI features (requires `ANTHROPIC_API_KEY`)
+
+### Changed
+- Updated documentation with glossary CLI commands
+- Added glossary to knowledge store structure
+
+## [0.2.0] - 2025-01-01
+
+### Added
+- REST API server with Express.js
+- SQLite adapter
+- MySQL/MariaDB adapter
+- Feedback loop with query correction tracking
+- Metrics collection and reporting
+- Cost estimation with EXPLAIN analysis
+- Auto-harvesting of DB metadata
+
 ## [0.1.0] - 2025-01-01
 
 ### Added