diff --git a/README.en.md b/README.en.md index 82a1663..a869a36 100644 --- a/README.en.md +++ b/README.en.md @@ -4,13 +4,17 @@ English | **[中文](README.md)** | **[日本語](README.ja.md)** --- -**The World's Most Comprehensive, Authoritative, and Structured Open Data Source Repository** +**The World's Most Comprehensive, Authoritative, and Structured Open Data Source Repository — Agent First** + +> **Agent First**: FirstData is designed with AI Agents as the primary user. Agents can automatically register, activate, and configure MCP via standardized Skills — zero human intervention required. [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![Data Sources](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/sources-count.json)](firstdata/indexes/statistics.json) [![Progress](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/progress.json)](firstdata/indexes/statistics.json) [![Authority](https://img.shields.io/badge/Authority-Government%20%26%20International%20First-brightgreen.svg)](#) [![MCP Server](https://img.shields.io/badge/MCP-AI%20Smart%20Search-purple.svg)](https://firstdata.deepminer.com.cn/) +[![Skill](https://img.shields.io/badge/Skill-Agent%20First-blue.svg)](skills/firstdata/SKILL.md) +[![ClawHub](https://img.shields.io/badge/ClawHub-firstdata-orange.svg)](https://clawhub.ai/ningzimu/firstdata) --- @@ -61,6 +65,7 @@ We systematically discover and aggregate high-trust sources across domains—cov | ⭐ **Authority Level Classification** | Six authority levels: government, international organizations, research institutions, market, commercial, and others | Scientifically assess data source credibility, provide quality filtering basis for AI | | 🤖 **AI Smart Search** | LLM-driven data source query Agent that understands complex multi-dimensional queries | Get authoritative data sources through natural language, no manual filtering needed | | 🔌 **MCP Protocol Integration** | Provides standard MCP Server, integrable with Claude Desktop, Cline, and other AI applications | Enable any AI application to access the authoritative data source knowledge base | +| 🤖 **Agent Skill Distribution** | Standardized Skill definition — Agents can auto-register tokens, auto-configure MCP, zero human intervention | Agent First — Let Agents access authoritative data like a built-in capability | | 🌍 **Bilingual Support** | All metadata provided in both Chinese and English | Connect global data ecosystems, break language barriers | | 🔍 **100% Verification** | Every URL tested, every data source with complete documentation, every authority level with justification | Ensure data sources are genuinely available, avoid broken links and hallucinated citations | @@ -487,366 +492,35 @@ All Conclusions Traceable to Original Documents for Verification ## 🚀 Quick Start -### Integrate FirstData MCP - -**Configure FirstData MCP to your AI application for intelligent search of authoritative data sources and operation manual navigation** - -Supports multiple platforms: Claude Desktop, Cline (VS Code), Zed, Cursor, and all AI applications compatible with MCP protocol. - ---- - -#### Configuration Guide: Choose Based on Your Platform - -> **📝 Important Note** -> -> **Apply for API Key (Required)**: Before configuring the MCP server, please visit [FirstData API Application](https://firstdata.deepminer.com.cn/) to apply for a free API key. Replace all `` in the configuration examples below with your actual API key. +FirstData offers two onboarding paths: **AI Agent auto-onboarding** (recommended) and **manual MCP configuration**. --- -
-Claude Desktop - -**Manual JSON Configuration** - -1. **Find Configuration File**: - - - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json` - - Windows: `%APPDATA%\Claude\claude_desktop_config.json` - - Linux: `~/.config/Claude/claude_desktop_config.json` -2. **Add Configuration**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **Restart Claude Desktop** to apply configuration - -
- -
-Cline (VS Code Extension) - -1. **Open Cline MCP Settings**: - - - Open Cline in VS Code - - Click settings icon → **Advanced MCP settings** - - Or directly edit `cline_mcp_settings.json` -2. **Add Configuration**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - - > **Tip**: The `autoApprove` field allows commonly used read-only tools to execute automatically without confirmation each time. - > -3. **After saving configuration**, Cline will automatically load the MCP server - -Reference: [Cline MCP Configuration Docs](https://docs.cline.bot/mcp/configuring-mcp-servers) - -
- -
-Zed Editor - -1. **Create Configuration File**: - - - Create `.zed/settings.json` in project root directory - - Or use global configuration: `~/.config/zed/settings.json` -2. **Add Configuration**: - - ```json - { - "context_servers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - }, - "enabled": true - } - } - } - ``` - - > **Note**: Zed uses `context_servers` instead of `mcpServers` - > -3. **Reload Zed** or restart project to apply configuration - -
- -
-Cursor Editor - -1. **Open Cursor Settings**: - - - `Cmd/Ctrl + Shift + P` → Search "MCP Settings" - - Or go to `Cursor Settings` → `MCP` → `New MCP Server` -2. **Add Configuration**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **Restart Cursor** to load MCP server - -
- -
-Copilot / VS Code +### Path 1: AI Agent Auto-Onboarding (Recommended) -**Recommended Method: HTTP Server** - -1. Reference [VS Code MCP Configuration Guide](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) -2. Add configuration to VS Code MCP settings: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -**Using VS Code CLI:** +If you're using an AI Agent that supports Skills (e.g., Claude Code + OpenClaw), install the FirstData Skill with one command: ```bash -code --add-mcp '{"name":"firstdata","type":"streamable-http","url":"https://firstdata.deepminer.com.cn/mcp","headers":{"Authorization":"Bearer "}}' -``` - -
- -
-Copilot CLI - -Use HTTP server method to connect: - -```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } -``` - -
- -
-Windsurf - -1. **Open Windsurf MCP Configuration**: - - - Reference [Windsurf MCP Configuration Guide](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) -2. **Add Configuration**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-JetBrains AI Assistant & Junie - -1. **Open JetBrains IDE Settings**: - - - Go to `Settings | Tools | AI Assistant | Model Context Protocol (MCP)` - - Or `Settings | Tools | Junie | MCP Settings` -2. **Click `Add` and add the following configuration**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-Warp Terminal - -1. **Open Warp Settings**: - - - Go to `Settings | AI | Manage MCP Servers` -2. **Click `+ Add` to add MCP server**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -Reference: [Warp MCP Configuration Docs](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) - -
- -
-Gemini CLI - -Reference [Gemini CLI MCP Configuration Guide](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md), use the following configuration: - -```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } +clawhub install firstdata ``` -
- -
-Gemini Code Assist - -Reference [Gemini Code Assist MCP Configuration Guide](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers), use the following configuration: +Once installed, the Agent can automatically register, activate, and configure MCP — no human action needed. -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` +> Skill definition: [`skills/firstdata/SKILL.md`](skills/firstdata/SKILL.md) | ClawHub page: [clawhub.ai/ningzimu/firstdata](https://clawhub.ai/ningzimu/firstdata) -
- -
-Factory CLI (Droid) - -Reference [Factory CLI MCP Configuration Docs](https://docs.factory.ai/cli/configuration/mcp), use the following configuration: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Qoder & Qoder CLI - -1. Open **Qoder Settings** -2. Go to `MCP Server` → `+ Add` -3. Add the following configuration: - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -Reference: [Qoder MCP Configuration Docs](https://docs.qoder.com/user-guide/chat/model-context-protocol) - -
+--- -
-Kiro +### Path 2: Manual MCP Configuration -**Method 1: Via Kiro Settings** +For manual setup. Supports Claude Desktop, Cline, Cursor, Zed, and all MCP-compatible AI applications. -1. Open **Kiro Settings** -2. Go to `Configure MCP` → `Open Workspace or User MCP Config` +#### Step 1: Get an API Key -**Method 2: Via Activity Bar** +Visit [FirstData API Application](https://firstdata.deepminer.com.cn/) to apply for a free API Key. -1. Select `Kiro` from IDE **Activity Bar** -2. Go to `MCP Servers` → `Click Open MCP Config` +#### Step 2: Add MCP Configuration -**Configuration Content:** +The core configuration is the same for all platforms. Add the following JSON to your platform's MCP config file: ```json { @@ -862,96 +536,32 @@ Reference: [Qoder MCP Configuration Docs](https://docs.qoder.com/user-guide/chat } ``` -
+> **Note**: Zed uses `context_servers` instead of `mcpServers` as the top-level key. -
-OpenCode - -1. **Create or edit configuration file**: - - - Path: `~/.config/opencode/opencode.json` - - Create it if the file doesn't exist -2. **Add the following configuration**: - - ```json - { - "$schema": "https://opencode.ai/config.json", - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -Reference: [OpenCode MCP Configuration Docs](https://opencode.ai/docs/mcp-servers) - -
+#### Step 3: Find Your Platform's Config File
-Visual Studio - -Reference Visual Studio MCP configuration documentation, use the following configuration: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Codex - -Reference [Codex MCP Configuration Guide](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp), use the following configuration: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Amp - -Reference [Amp MCP Configuration Docs](https://ampcode.com/manual#mcp), use the following configuration: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "http", - "url": "http://localhost:8001/mcp", - "headers": { - "Authorization": "Bearer your_mcp_api_key_here" - } - } - } -} -``` +Platform config file locations + +| Platform | Config File Location | Reference | +|----------|---------------------|-----------| +| **Claude Desktop** | macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
Windows: `%APPDATA%\Claude\claude_desktop_config.json`
Linux: `~/.config/Claude/claude_desktop_config.json` | — | +| **Cline (VS Code)** | Cline → Settings → Advanced MCP settings, or `cline_mcp_settings.json` | [Docs](https://docs.cline.bot/mcp/configuring-mcp-servers) | +| **Cursor** | `Cmd/Ctrl+Shift+P` → "MCP Settings", or Cursor Settings → MCP | — | +| **Copilot / VS Code** | VS Code MCP settings, or CLI: `code --add-mcp '{...}'` | [Docs](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) | +| **Zed** | `.zed/settings.json` or `~/.config/zed/settings.json` (uses `context_servers`) | — | +| **JetBrains** | Settings → Tools → AI Assistant → MCP, or Junie → MCP Settings | — | +| **Windsurf** | Windsurf MCP config | [Docs](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) | +| **Warp** | Settings → AI → Manage MCP Servers | [Docs](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) | +| **Gemini CLI** | Gemini CLI MCP config | [Docs](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md) | +| **Gemini Code Assist** | Gemini Code Assist MCP config | [Docs](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers) | +| **Factory CLI (Droid)** | Factory CLI config | [Docs](https://docs.factory.ai/cli/configuration/mcp) | +| **Qoder** | Qoder Settings → MCP Server → + Add | [Docs](https://docs.qoder.com/user-guide/chat/model-context-protocol) | +| **Kiro** | Kiro Settings → Configure MCP, or Activity Bar → MCP Servers | — | +| **OpenCode** | `~/.config/opencode/opencode.json` | [Docs](https://opencode.ai/docs/mcp-servers) | +| **Visual Studio** | Visual Studio MCP config | — | +| **Codex** | Codex MCP config | [Docs](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp) | +| **Amp** | Amp MCP config | [Docs](https://ampcode.com/manual#mcp) |
diff --git a/README.ja.md b/README.ja.md index ed608d0..deb5fc3 100644 --- a/README.ja.md +++ b/README.ja.md @@ -4,13 +4,17 @@ --- -**世界最も包括的・権威的・構造化されたオープンデータソースリポジトリ** +**世界最も包括的・権威的・構造化されたオープンデータソースリポジトリ — Agent First** + +> **Agent First**:FirstData は AI Agent を第一優先ユーザーとして設計されています。Agent は標準化された Skill を通じて登録・アクティベーション・MCP 設定を自動で完了でき、人手を介する必要はありません。 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![Data Sources](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/sources-count.json)](firstdata/indexes/statistics.json) [![Progress](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/progress.json)](firstdata/indexes/statistics.json) [![Authority](https://img.shields.io/badge/Authority-Government%20%26%20International%20First-brightgreen.svg)](#) [![MCP Server](https://img.shields.io/badge/MCP-AI%20Smart%20Search-purple.svg)](https://firstdata.deepminer.com.cn/) +[![Skill](https://img.shields.io/badge/Skill-Agent%20First-blue.svg)](skills/firstdata/SKILL.md) +[![ClawHub](https://img.shields.io/badge/ClawHub-firstdata-orange.svg)](https://clawhub.ai/ningzimu/firstdata) --- @@ -61,6 +65,7 @@ | ⭐ **権威レベル分類** | 政府、国際機関、研究機関、市場、商業、その他の6つの権威レベル | データソースの信頼性を科学的に評価し、AIの品質フィルタリング基準を提供 | | 🤖 **AIスマート検索** | 複雑な多次元クエリを理解するLLM駆動のデータソースクエリエージェント | 自然言語で権威あるデータソースを取得し、手動フィルタリング不要 | | 🔌 **MCPプロトコル統合** | 標準MCPサーバーを提供、Claude Desktop、Clineなどのアプリケーションと統合可能 | 任意のAIアプリケーションが権威あるデータソースのナレッジベースにアクセス可能 | +| 🤖 **Agent Skill 配信** | 標準化された Skill 定義 — Agent が自動でトークン登録・MCP設定を完了、人手不要 | Agent First — Agent が組み込み機能のように権威データにアクセス | | 🌍 **バイリンガルサポート** | すべてのメタデータを中国語と英語で提供 | グローバルなデータエコシステムを繋ぎ、言語の壁を打ち破る | | 🔍 **100%検証** | すべてのURLをテスト済み、すべてのデータソースに完全な文書、すべての権威レベルに根拠あり | データソースが本当に利用可能であることを確保し、リンク切れや幻覚的な引用を回避 | @@ -487,366 +492,35 @@ MCPスマート検索が権威あるデータソースを推奨(HKEX) ## 🚀 クイックスタート -### FirstData MCPの統合 - -**FirstData MCPをAIアプリケーションに設定し、権威あるデータソースのインテリジェント検索と操作マニュアルナビゲーションを利用しましょう** - -複数のプラットフォームをサポート:Claude Desktop、Cline(VS Code)、Zed、Cursor、およびMCPプロトコルと互換性のあるすべてのAIアプリケーション。 - ---- - -#### 設定ガイド:お使いのプラットフォームに応じて選択 - -> **📝 重要なメモ** -> -> **APIキーの申請(必須)**: MCPサーバーを設定する前に、[FirstData API申請](https://firstdata.deepminer.com.cn/) にアクセスして無料のAPIキーを申請してください。以下の設定例の `` をすべて実際のAPIキーに置き換えてください。 +FirstData は2つのオンボーディングパスを提供します:**AI Agent 自動オンボーディング**(推奨)と**手動 MCP 設定**。 --- -
-Claude Desktop - -**手動JSON設定** - -1. **設定ファイルの場所を確認**: - - - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json` - - Windows: `%APPDATA%\Claude\claude_desktop_config.json` - - Linux: `~/.config/Claude/claude_desktop_config.json` -2. **設定を追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **Claude Desktopを再起動**して設定を適用 - -
- -
-Cline (VS Code拡張機能) - -1. **Cline MCP設定を開く**: - - - VS CodeでClineを開く - - 設定アイコンをクリック → **高度なMCP設定** - - または直接 `cline_mcp_settings.json` を編集 -2. **設定を追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - - > **ヒント**: `autoApprove` フィールドを使用すると、よく使われる読み取り専用ツールが毎回確認なしに自動実行されます。 - > -3. **設定を保存後**、ClineがMCPサーバーを自動的にロード - -参考: [Cline MCP設定ドキュメント](https://docs.cline.bot/mcp/configuring-mcp-servers) - -
- -
-Zed Editor - -1. **設定ファイルを作成**: - - - プロジェクトのルートディレクトリに `.zed/settings.json` を作成 - - またはグローバル設定を使用: `~/.config/zed/settings.json` -2. **設定を追加**: - - ```json - { - "context_servers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - }, - "enabled": true - } - } - } - ``` - - > **注意**: Zedは `mcpServers` の代わりに `context_servers` を使用 - > -3. **Zedをリロード**またはプロジェクトを再起動して設定を適用 - -
- -
-Cursor Editor - -1. **Cursor設定を開く**: - - - `Cmd/Ctrl + Shift + P` → "MCP Settings"を検索 - - または `Cursor Settings` → `MCP` → `New MCP Server` に移動 -2. **設定を追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **Cursorを再起動**してMCPサーバーをロード - -
- -
-Copilot / VS Code +### パス1:AI Agent 自動オンボーディング(推奨) -**推奨方法: HTTPサーバー** - -1. [VS Code MCP設定ガイド](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) を参照 -2. VS Code MCP設定に以下の設定を追加: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -**VS Code CLIを使用:** +Skill をサポートする AI Agent(Claude Code + OpenClaw など)を使用している場合、ワンコマンドで FirstData Skill をインストールできます: ```bash -code --add-mcp '{"name":"firstdata","type":"streamable-http","url":"https://firstdata.deepminer.com.cn/mcp","headers":{"Authorization":"Bearer "}}' -``` - -
- -
-Copilot CLI - -HTTPサーバー方式を使用して接続: - -```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } -``` - -
- -
-Windsurf - -1. **Windsurf MCP設定を開く**: - - - [Windsurf MCP設定ガイド](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) を参照 -2. **設定を追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-JetBrains AI Assistant & Junie - -1. **JetBrains IDE設定を開く**: - - - `Settings | Tools | AI Assistant | Model Context Protocol (MCP)` に移動 - - または `Settings | Tools | Junie | MCP Settings` -2. **`Add` をクリックして以下の設定を追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-Warp Terminal - -1. **Warp設定を開く**: - - - `Settings | AI | Manage MCP Servers` に移動 -2. **`+ Add` をクリックしてMCPサーバーを追加**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考: [Warp MCP設定ドキュメント](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) - -
- -
-Gemini CLI - -[Gemini CLI MCP設定ガイド](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md) を参照し、以下の設定を使用: - -```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } +clawhub install firstdata ``` -
- -
-Gemini Code Assist - -[Gemini Code Assist MCP設定ガイド](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers) を参照し、以下の設定を使用: +インストール後、Agent が登録・アクティベーション・MCP 設定を**自動で完了**します。人手は不要です。 -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` +> Skill 定義ファイル:[`skills/firstdata/SKILL.md`](skills/firstdata/SKILL.md) | ClawHub ページ:[clawhub.ai/ningzimu/firstdata](https://clawhub.ai/ningzimu/firstdata) -
- -
-Factory CLI (Droid) - -[Factory CLI MCP設定ドキュメント](https://docs.factory.ai/cli/configuration/mcp) を参照し、以下の設定を使用: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Qoder & Qoder CLI - -1. **Qoder設定**を開く -2. `MCP Server` → `+ Add` に移動 -3. 以下の設定を追加: - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考: [Qoder MCP設定ドキュメント](https://docs.qoder.com/user-guide/chat/model-context-protocol) - -
+--- -
-Kiro +### パス2:手動 MCP 設定 -**方法1:Kiro設定経由** +手動設定が必要な場合。Claude Desktop、Cline、Cursor、Zed など、MCP 対応のすべての AI アプリケーションをサポートします。 -1. **Kiro設定**を開く -2. `Configure MCP` → `Open Workspace or User MCP Config` に移動 +#### ステップ1:API キーの取得 -**方法2:アクティビティバー経由** +[FirstData API 申請](https://firstdata.deepminer.com.cn/) にアクセスして無料の API キーを申請してください。 -1. IDE**アクティビティバー**から `Kiro` を選択 -2. `MCP Servers` → `Click Open MCP Config` に移動 +#### ステップ2:MCP 設定の追加 -**設定内容:** +すべてのプラットフォームで共通の設定です。以下の JSON をお使いのプラットフォームの MCP 設定ファイルに追加してください: ```json { @@ -862,96 +536,32 @@ HTTPサーバー方式を使用して接続: } ``` -
+> **注意**:Zed はトップレベルキーとして `mcpServers` ではなく `context_servers` を使用します。 -
-OpenCode - -1. **設定ファイルを作成または編集**: - - - パス: `~/.config/opencode/opencode.json` - - ファイルが存在しない場合は作成 -2. **以下の設定を追加**: - - ```json - { - "$schema": "https://opencode.ai/config.json", - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考: [OpenCode MCP設定ドキュメント](https://opencode.ai/docs/mcp-servers) - -
+#### ステップ3:プラットフォームの設定ファイルを見つける
-Visual Studio - -Visual Studio MCP設定ドキュメントを参照し、以下の設定を使用: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Codex - -[Codex MCP設定ガイド](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp) を参照し、以下の設定を使用: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Amp - -[Amp MCP設定ドキュメント](https://ampcode.com/manual#mcp) を参照し、以下の設定を使用: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "http", - "url": "http://localhost:8001/mcp", - "headers": { - "Authorization": "Bearer your_mcp_api_key_here" - } - } - } -} -``` +各プラットフォームの設定ファイル位置 + +| プラットフォーム | 設定ファイルの場所 | 参考ドキュメント | +|----------|---------------------|-----------| +| **Claude Desktop** | macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
Windows: `%APPDATA%\Claude\claude_desktop_config.json`
Linux: `~/.config/Claude/claude_desktop_config.json` | — | +| **Cline (VS Code)** | Cline → 設定 → Advanced MCP settings、または `cline_mcp_settings.json` | [ドキュメント](https://docs.cline.bot/mcp/configuring-mcp-servers) | +| **Cursor** | `Cmd/Ctrl+Shift+P` → "MCP Settings"、または Cursor Settings → MCP | — | +| **Copilot / VS Code** | VS Code MCP 設定、または CLI:`code --add-mcp '{...}'` | [ドキュメント](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) | +| **Zed** | `.zed/settings.json` または `~/.config/zed/settings.json`(`context_servers` を使用) | — | +| **JetBrains** | Settings → Tools → AI Assistant → MCP、または Junie → MCP Settings | — | +| **Windsurf** | Windsurf MCP 設定 | [ドキュメント](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) | +| **Warp** | Settings → AI → Manage MCP Servers | [ドキュメント](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) | +| **Gemini CLI** | Gemini CLI MCP 設定 | [ドキュメント](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md) | +| **Gemini Code Assist** | Gemini Code Assist MCP 設定 | [ドキュメント](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers) | +| **Factory CLI (Droid)** | Factory CLI 設定 | [ドキュメント](https://docs.factory.ai/cli/configuration/mcp) | +| **Qoder** | Qoder Settings → MCP Server → + Add | [ドキュメント](https://docs.qoder.com/user-guide/chat/model-context-protocol) | +| **Kiro** | Kiro Settings → Configure MCP、または Activity Bar → MCP Servers | — | +| **OpenCode** | `~/.config/opencode/opencode.json` | [ドキュメント](https://opencode.ai/docs/mcp-servers) | +| **Visual Studio** | Visual Studio MCP 設定 | — | +| **Codex** | Codex MCP 設定 | [ドキュメント](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp) | +| **Amp** | Amp MCP 設定 | [ドキュメント](https://ampcode.com/manual#mcp) |
diff --git a/README.md b/README.md index 7715289..380d686 100644 --- a/README.md +++ b/README.md @@ -4,15 +4,19 @@ --- -**全球最全面、最权威、最结构化的开源数据源知识库** +**全球最全面、最权威、最结构化的开源数据源知识库 — Agent First** **The World's Most Comprehensive, Authoritative, and Structured Open Data Source Repository** +> **Agent First**:FirstData 以 AI Agent 为第一优先用户设计。Agent 可通过标准化 Skill 自动完成注册、激活和 MCP 配置,零人工介入即可接入权威数据源知识库。 + [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![Data Sources](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/sources-count.json)](firstdata/indexes/statistics.json) [![Progress](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/assets/badges/progress.json)](firstdata/indexes/statistics.json) [![Authority](https://img.shields.io/badge/Authority-Government%20%26%20International%20First-brightgreen.svg)](#) [![MCP Server](https://img.shields.io/badge/MCP-AI%20Smart%20Search-purple.svg)](https://firstdata.deepminer.com.cn/) +[![Skill](https://img.shields.io/badge/Skill-Agent%20First-blue.svg)](skills/firstdata/SKILL.md) +[![ClawHub](https://img.shields.io/badge/ClawHub-firstdata-orange.svg)](https://clawhub.ai/ningzimu/firstdata) --- @@ -63,6 +67,7 @@ | ⭐**权威等级分类** | 政府、国际组织、研究机构、市场、商业等六类权威等级 | 科学评估数据源可信度,为AI提供质量过滤依据 | | 🤖**AI智能搜索** | 基于LLM驱动的数据源查询Agent,理解复杂多维度查询 | 自然语言即可获取权威数据源,无需人工筛选 | | 🔌**MCP协议集成** | 提供标准MCP Server,可集成到Claude Desktop、Cline等AI应用 | 让任何AI应用都能访问权威数据源知识库 | +| 🤖**Agent Skill 分发** | 标准化 Skill 定义,Agent 可自动注册 token、自动配置 MCP,零人工介入 | Agent First — 让 Agent 像调用内置能力一样接入权威数据 | | 🌍**中英双语支持** | 所有元数据提供中英文版本 | 连接全球数据生态,打破语言壁垒 | | 🔍**100%验证** | 每个URL经过测试,每个数据源有完整文档,每个权威等级有依据 | 确保数据源真实可用,避免断链和幻觉引用 | @@ -488,366 +493,35 @@ MCP 智能检索推荐权威数据源(港交所披露易) ## 🚀 快速开始 -### 集成 FirstData MCP - -**将FirstData MCP配置到您的AI应用中,实现智能检索权威数据源和操作说明书导航** - -支持多个平台:Claude Desktop、Cline (VS Code)、Zed、Cursor 等所有兼容 MCP 协议的 AI 应用。 - ---- - -#### 配置指南:根据你使用的平台选择 - -> **📝 重要提示** -> -> **申请 API Key(必需)**:在配置 MCP 服务器之前,请先访问 [FirstData API 申请](https://firstdata.deepminer.com.cn/) 申请免费的 API key。将下方所有配置示例中的 `` 替换为你申请到的实际 API key。 +FirstData 提供两条接入路径:**AI Agent 自动接入**(推荐)和**手动配置 MCP**。 --- -
-Claude Desktop - -**手动 JSON 配置** - -1. **找到配置文件**: - - - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json` - - Windows: `%APPDATA%\Claude\claude_desktop_config.json` - - Linux: `~/.config/Claude/claude_desktop_config.json` -2. **添加配置**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **重启 Claude Desktop** 使配置生效 - -
- -
-Cline (VS Code 扩展) - -1. **打开 Cline MCP 设置**: - - - 在 VS Code 中打开 Cline - - 点击设置图标 → **Advanced MCP settings** - - 或直接编辑 `cline_mcp_settings.json` -2. **添加配置**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - - > **提示**:`autoApprove` 字段可以让常用的只读工具自动执行,无需每次确认。 - > -3. **保存配置**后,Cline 会自动加载 MCP 服务器 - -参考:[Cline MCP 配置文档](https://docs.cline.bot/mcp/configuring-mcp-servers) - -
- -
-Zed 编辑器 - -1. **创建配置文件**: - - - 在项目根目录创建 `.zed/settings.json` - - 或使用全局配置:`~/.config/zed/settings.json` -2. **添加配置**: - - ```json - { - "context_servers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - }, - "enabled": true - } - } - } - ``` - - > **注意**:Zed 使用 `context_servers` 而不是 `mcpServers` - > -3. **重新加载 Zed** 或重启项目以应用配置 - -
- -
-Cursor 编辑器 - -1. **打开 Cursor 设置**: - - - `Cmd/Ctrl + Shift + P` → 搜索 "MCP Settings" - - 或进入 `Cursor Settings` → `MCP` → `New MCP Server` -2. **添加配置**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` -3. **重启 Cursor** 以加载 MCP 服务器 - -
- -
-Copilot / VS Code - -**推荐方式:HTTP 服务器** +### 路径一:AI Agent 自动接入(推荐) -1. 参考 [VS Code MCP 配置指南](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) -2. 添加配置到 VS Code MCP 设置: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -**使用 VS Code CLI:** +如果你正在使用支持 Skill 的 AI Agent(如 Claude Code + OpenClaw 等),可以一键安装 FirstData Skill: ```bash -code --add-mcp '{"name":"firstdata","type":"streamable-http","url":"https://firstdata.deepminer.com.cn/mcp","headers":{"Authorization":"Bearer "}}' -``` - -
- -
-Copilot CLI - -使用 HTTP 服务器方式连接: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Windsurf - -1. **打开 Windsurf MCP 配置**: - - - 参考 [Windsurf MCP 配置指南](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) -2. **添加配置**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-JetBrains AI Assistant & Junie - -1. **打开 JetBrains IDE 设置**: - - - 进入 `Settings | Tools | AI Assistant | Model Context Protocol (MCP)` - - 或 `Settings | Tools | Junie | MCP Settings` -2. **点击 `Add` 并添加以下配置**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -
- -
-Warp Terminal - -1. **打开 Warp 设置**: - - - 进入 `Settings | AI | Manage MCP Servers` -2. **点击 `+ Add` 添加 MCP 服务器**: - - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考:[Warp MCP 配置文档](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) - -
- -
-Gemini CLI - -参考 [Gemini CLI MCP 配置指南](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md),使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Gemini Code Assist - -参考 [Gemini Code Assist MCP 配置指南](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers),使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Factory CLI (Droid) - -参考 [Factory CLI MCP 配置文档](https://docs.factory.ai/cli/configuration/mcp),使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} +clawhub install firstdata ``` -
+安装后 Agent 可自动完成注册、激活和 MCP 配置,无需人工操作。 -
-Qoder & Qoder CLI - -1. 打开 **Qoder Settings** -2. 进入 `MCP Server` → `+ Add` -3. 添加以下配置: - ```json - { - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考:[Qoder MCP 配置文档](https://docs.qoder.com/user-guide/chat/model-context-protocol) +> Skill 定义文件:[`skills/firstdata/SKILL.md`](skills/firstdata/SKILL.md) | ClawHub 页面:[clawhub.ai/ningzimu/firstdata](https://clawhub.ai/ningzimu/firstdata) -
+--- -
-Kiro +### 路径二:手动配置 MCP -**方式一:通过 Kiro Settings** +适用于人工配置场景。支持 Claude Desktop、Cline、Cursor、Zed 等所有兼容 MCP 协议的 AI 应用。 -1. 打开 **Kiro Settings** -2. 进入 `Configure MCP` → `Open Workspace or User MCP Config` +#### 第 1 步:获取 API Key -**方式二:通过 Activity Bar** +访问 [FirstData API 申请](https://firstdata.deepminer.com.cn/) 申请免费的 API Key。 -1. 从 IDE **Activity Bar** 选择 `Kiro` -2. 进入 `MCP Servers` → `Click Open MCP Config` +#### 第 2 步:添加 MCP 配置 -**配置内容:** +所有平台的核心配置相同,将以下 JSON 添加到你所用平台的 MCP 配置文件中: ```json { @@ -863,96 +537,32 @@ code --add-mcp '{"name":"firstdata","type":"streamable-http","url":"https://firs } ``` -
+> **注意**:Zed 使用 `context_servers` 而非 `mcpServers` 作为顶层 key。 -
-OpenCode - -1. **创建或编辑配置文件**: - - - 路径:`~/.config/opencode/opencode.json` - - 如果文件不存在,请创建它 -2. **添加以下配置**: - - ```json - { - "$schema": "https://opencode.ai/config.json", - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } - } - ``` - -参考:[OpenCode MCP 配置文档](https://opencode.ai/docs/mcp-servers) - -
+#### 第 3 步:找到你的平台配置文件
-Visual Studio - -参考 Visual Studio MCP 配置文档,使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Codex - -参考 [Codex MCP 配置指南](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp),使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` - -
- -
-Amp - -参考 [Amp MCP 配置文档](https://ampcode.com/manual#mcp),使用以下配置: - -```json -{ - "mcpServers": { - "firstdata": { - "type": "streamable-http", - "url": "https://firstdata.deepminer.com.cn/mcp", - "headers": { - "Authorization": "Bearer " - } - } - } -} -``` +各平台配置文件位置 + +| 平台 | 配置文件位置 | 参考文档 | +|------|-------------|---------| +| **Claude Desktop** | macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
Windows: `%APPDATA%\Claude\claude_desktop_config.json`
Linux: `~/.config/Claude/claude_desktop_config.json` | — | +| **Cline (VS Code)** | Cline → 设置 → Advanced MCP settings,或 `cline_mcp_settings.json` | [文档](https://docs.cline.bot/mcp/configuring-mcp-servers) | +| **Cursor** | `Cmd/Ctrl+Shift+P` → "MCP Settings",或 Cursor Settings → MCP | — | +| **Copilot / VS Code** | VS Code MCP 设置,或 CLI:`code --add-mcp '{...}'` | [文档](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server) | +| **Zed** | `.zed/settings.json` 或 `~/.config/zed/settings.json`(使用 `context_servers`) | — | +| **JetBrains** | Settings → Tools → AI Assistant → MCP,或 Junie → MCP Settings | — | +| **Windsurf** | Windsurf MCP 配置 | [文档](https://docs.windsurf.com/windsurf/cascade/mcp#mcp-config-json) | +| **Warp** | Settings → AI → Manage MCP Servers | [文档](https://docs.warp.dev/knowledge-and-collaboration/mcp#adding-an-mcp-server) | +| **Gemini CLI** | Gemini CLI MCP 配置 | [文档](https://github.com/google-gemini/gemini-cli/blob/main/docs/tools/mcp-server.md) | +| **Gemini Code Assist** | Gemini Code Assist MCP 配置 | [文档](https://cloud.google.com/gemini/docs/codeassist/use-agentic-chat-pair-programmer#configure-mcp-servers) | +| **Factory CLI (Droid)** | Factory CLI 配置 | [文档](https://docs.factory.ai/cli/configuration/mcp) | +| **Qoder** | Qoder Settings → MCP Server → + Add | [文档](https://docs.qoder.com/user-guide/chat/model-context-protocol) | +| **Kiro** | Kiro Settings → Configure MCP,或 Activity Bar → MCP Servers | — | +| **OpenCode** | `~/.config/opencode/opencode.json` | [文档](https://opencode.ai/docs/mcp-servers) | +| **Visual Studio** | Visual Studio MCP 配置 | — | +| **Codex** | Codex MCP 配置 | [文档](https://github.com/openai/codex/blob/main/docs/advanced.md#model-context-protocol-mcp) | +| **Amp** | Amp MCP 配置 | [文档](https://ampcode.com/manual#mcp) |
diff --git a/skills/firstdata/SKILL.md b/skills/firstdata/SKILL.md new file mode 100644 index 0000000..498f608 --- /dev/null +++ b/skills/firstdata/SKILL.md @@ -0,0 +1,95 @@ +--- +name: firstdata +description: Find official portals, APIs, and download paths for authoritative primary data sources (governments, international organizations, research institutions, etc.). Use when users need to know "where to find this data from an official source", "which source is more authoritative", or "how to cite primary data". Covers 1000+ global data sources with authority comparison and site navigation guidance. +version: 0.0.2 +metadata: + openclaw: + homepage: https://github.com/MLT-OSS/FirstData + primaryEnv: FIRSTDATA_API_KEY + emoji: "📊" + requires: + env: + - FIRSTDATA_API_KEY + bins: + - curl +--- +# FirstData + +## What FirstData Is + +FirstData is a structured knowledge base of authoritative primary data sources, covering 1000+ sources to help agents locate official origins rather than generating unverified answers. + +It does not replace raw data — it acts as an "authoritative data navigator", taking vague user needs as input, recommending the most appropriate primary sources, and providing clear access paths, API information, and download methods so both users and agents can trace back to original evidence. + +**Coverage**: + +- International organizations: World Bank, IMF, OECD, WHO, FAO, etc. +- Chinese government agencies: PBC, National Bureau of Statistics, General Administration of Customs, CSRC, etc. +- National official agencies: US, Canada, Japan, UK, Australia, etc. +- Academic & research databases: NBER, Penn World Table, PubMed, etc. +- Corporate disclosure & market platforms: stock exchange disclosure systems, listed company filings, etc. +- Industry-specific databases: energy, finance, health, climate, legal & regulatory, etc. + +**When to use**: When users need to find official data sources, compare source authority, obtain official URLs/APIs/download paths, or build evidence-chain workflows. FirstData is a source locator, not an answer generator — after receiving results, guide users back to original sources for verification rather than treating them as final answers. + +## Capabilities + +**1. Source Locator** — Returns the top 3–5 most relevant sources with authority level, matching rationale, access URL, API documentation, and download methods. + +**2. Site Pathfinder** — Provides step-by-step navigation from homepage to target data for complex official websites, including alternative paths and API access methods. + +**3. Evidence-Ready Workflows** — Can be embedded into workflows requiring evidence chains: deep research, policy analysis, investment research, compliance auditing, fact-checking, etc. + +Each data source includes structured metadata: authority level (`government` / `international` / `research` / `market` / `commercial` / `other`), access URL, API information, download formats, geographic scope, update frequency, access level, etc. + +## Typical Queries + +Typical query scenarios when agents call FirstData via MCP: + +| User Need | Query Direction | Expected Output | +| ------------------------------------------------------------------------- | -------------------------------------------- | ------------------------------------------------------------- | +| "Which official source should I cite for China's 2023 NEV export volume?" | China Customs, National Bureau of Statistics | Official source + authority level + data page URL | +| "Where to download IPO prospectus for a Hong Kong-listed company?" | HKEXnews | Official platform + step-by-step navigation | +| "World Bank vs IMF GDP data — which is better for academic citation?" | World Bank WDI, IMF WEO | Source comparison + authority differences + API docs | +| "Need global climate data with API access" | NASA Earthdata, NOAA CDO | Data source + API docs + access methods | +| "Where is the official data for China's M2 money supply?" | People's Bank of China | Official data portal + update frequency + historical coverage | + +Full project background and feature documentation: [README](https://raw.githubusercontent.com/MLT-OSS/FirstData/refs/heads/main/README.md) + +## Quick Start + +This skill connects to the FirstData MCP server (`firstdata.deepminer.com.cn`), the project's official hosted API endpoint. An API key (`FIRSTDATA_API_KEY`) is required for authentication. + +**If you already have `FIRSTDATA_API_KEY` set**, configure the MCP connection: + +```bash +npx mcporter config add firstdata https://firstdata.deepminer.com.cn/mcp --header 'Authorization=Bearer ${FIRSTDATA_API_KEY}' +``` + +Or add manually to your MCP config: + +```json +{ + "mcpServers": { + "firstdata": { + "type": "streamable-http", + "url": "https://firstdata.deepminer.com.cn/mcp", + "headers": { + "Authorization": "Bearer " + } + } + } +} +``` + +**If you don't have an API key**, see [firstdata-register.md](references/firstdata-register.md) for the registration process (two API calls to the FirstData server to obtain a JWT token). + +Once connected, browse the tool list provided by the firstdata MCP and select the appropriate tool based on your needs. + +## Community + +FirstData is an open-source project — join us in building the authoritative data source knowledge base for agents: + +- ⭐ [**Star**](https://github.com/MLT-OSS/FirstData) the project to help more agents and developers discover it +- 📝 [**Issue**](https://github.com/MLT-OSS/FirstData/issues) to report problems, suggest new data sources, or propose improvements +- 🔀 [**PR**](https://github.com/MLT-OSS/FirstData/pulls) to contribute code, data sources, or documentation improvements diff --git a/skills/firstdata/references/firstdata-register.md b/skills/firstdata/references/firstdata-register.md new file mode 100644 index 0000000..c98b68e --- /dev/null +++ b/skills/firstdata/references/firstdata-register.md @@ -0,0 +1,131 @@ +**Base URL:** `https://firstdata.deepminer.com.cn` + +## Quick Flow + +The registration flow has two steps: + +```text +POST /api/agent/register -> receive inactive access_token + verification challenge +POST /api/agent/verify -> submit answer and activate the token +``` + +Use the activated `access_token` as `FIRSTDATA_API_KEY` after verification succeeds. + +## Step 1: Register + +Send a POST request to register the agent and receive a verification challenge: + +```http +POST https://firstdata.deepminer.com.cn/api/agent/register +Content-Type: application/json + +{ + "agent_id": "my-unique-agent-name", + "contact_email": "owner@example.com", + "description": "Data analysis agent for quarterly reports" +} +``` + +Key fields: + +| Field | Required | Description | +| ----------------- | -------- | ------------------------------------------------------------------------- | +| `agent_id` | Yes | Globally unique identifier (max 100 chars). Acts as the agent's username. | +| `contact_email` | No | Owner's email for contact purposes | +| `description` | Yes | Brief description of the agent's purpose (max 500 chars) | + +**Success response** (`success: true`): + +```json +{ + "success": true, + "data": { + "agent_id": "my-unique-agent-name", + "access_token": "eyJhbGciOi...", + "verification": { + "verification_code": "a1b2c3d4-...", + "challenge_text": "...", + "expires_at": "2025-01-15T10:05:00", + "instructions": "..." + } + } +} +``` + +Save both `access_token` and `verification.verification_code`. The token is returned immediately but is not usable until verification succeeds. + +## Step 2: Solve the Challenge and Verify + +Read `verification.challenge_text`, follow `verification.instructions`, solve the math challenge, and submit the answer: + +```http +POST https://firstdata.deepminer.com.cn/api/agent/verify +Content-Type: application/json + +{ + "verification_code": "a1b2c3d4-...", + "answer": "your answer here" +} +``` + +Success response: + +```json +{ + "success": true, + "message": "验证成功,token 已激活", + "mcp_server_url": "https://mcp.firstdata.deepminer.com.cn" +} +``` + +After this step, the `access_token` from Step 1 becomes active. If `mcp_server_url` is returned, use it as the MCP endpoint. + +## After Activation + +- Use the activated `access_token` as `FIRSTDATA_API_KEY` +- Connect with `Authorization: Bearer ` +- The token is a JWT and is valid for **365 days** by default + +Rate limits: + +- **RPS limit:** 10 requests per second +- **Daily quota:** 5000 requests per day + +## Error Handling + +| Scenario | What happens | Action | +| ------------------------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------------------ | +| `agent_id` already registered with active token | Registration returns `success: false` | Choose a different `agent_id` or wait for the existing token to expire | +| Pending verification exists for this `agent_id` | Registration returns `success: false` | Complete the existing verification or wait for it to expire | +| Wrong answer | Verify returns `success: false` with remaining attempts | Re-read the challenge carefully and try again | +| All attempts exhausted | Verify returns `success: false`, status becomes `failed` | Start over with a new `/api/agent/register` call | +| Challenge expired | Verify returns `success: false` | Start over with a new `/api/agent/register` call | +| IP rate limited | HTTP 429 | Wait before retrying | + +## MCP Configuration + +After activation, configure the MCP connection using either method: + +**Option 1: MCPorter CLI (recommended)** + +```bash +npx mcporter config add firstdata https://firstdata.deepminer.com.cn/mcp --header 'Authorization=Bearer ${FIRSTDATA_API_KEY}' +``` + +**Option 2: Manual MCP configuration** + +```json +{ + "mcpServers": { + "firstdata": { + "type": "streamable-http", + "url": "https://firstdata.deepminer.com.cn/mcp", + "headers": { + "Authorization": "Bearer " + } + } + } +} +``` + +`FIRSTDATA_API_KEY` should be set to the activated `access_token`.