Claude Code Gateway

Claude Code Gateway

Enables local Claude Code to interact with Snowflake, Asana, M365, GitHub, and Hive Mind with automatic response truncation to prevent memory overload.

Category
访问服务器

README

Claude Code Gateway - Response-Limited MCP Server

Overview

This is a response-limited version of the Sovereign Mind MCP Gateway, specifically designed for use with Claude Code on local machines. It prevents memory overload by automatically truncating large responses.

Key Features

  • Response Size Limiting: Max 50KB per response (configurable)
  • Row Limiting: Database queries limited to 100 rows by default
  • Pagination Hints: Automatically adds hints when data is truncated
  • Memory Safety: JSON serialization with overflow protection

Response Limits (Configurable)

Parameter Default Description
MAX_RESPONSE_BYTES 51200 Maximum response size in bytes
MAX_ROWS 100 Maximum rows from database queries
MAX_EMAIL_COUNT 25 Maximum emails per request
MAX_TASK_COUNT 50 Maximum Asana tasks per request
MAX_FILE_LIST 100 Maximum files in directory listings

Installation

Option 1: Local Python (Recommended for Claude Code)

# Clone or copy files
cd /path/to/claude-code-gateway

# Create virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Copy and configure environment
cp .env.template .env
# Edit .env with your credentials

Option 2: Docker (For Remote Access)

# Build image
docker build -t claude-code-gateway:latest .

# Run locally
docker run -p 8000:8000 --env-file .env claude-code-gateway:latest

Option 3: Azure Container Apps

# Build and push to ACR
az acr build --registry sovereignmindacr \
  --image claude-code-gateway:v1.0.0 \
  --file Dockerfile .

# Deploy to Container Apps
az containerapp update \
  --name claude-code-gateway \
  --resource-group SovereignMind-RG \
  --image sovereignmindacr.azurecr.io/claude-code-gateway:v1.0.0

Claude Code Configuration

Add to your Claude Code MCP configuration (~/.config/claude-code/mcp.json or equivalent):

{
  "mcpServers": {
    "sovereign-mind": {
      "command": "python",
      "args": ["/path/to/claude-code-gateway/gateway.py"],
      "env": {
        "SNOWFLAKE_PASSWORD": "your-password",
        "ASANA_TOKEN": "your-token",
        "GITHUB_TOKEN": "your-token",
        "M365_TENANT_ID": "your-tenant",
        "M365_CLIENT_ID": "your-client-id",
        "M365_CLIENT_SECRET": "your-secret"
      }
    }
  }
}

Available Tools

Gateway

  • gateway_status - Get current limits and configuration

Snowflake

  • sm_query_snowflake - Execute SQL with automatic row limiting

Hive Mind

  • hivemind_write - Write to shared memory
  • hivemind_read - Read from shared memory (limited)

Asana

  • asana_list_tasks - List tasks (limited to 50)
  • asana_create_task - Create a task
  • asana_get_task - Get task details
  • asana_complete_task - Mark task complete
  • asana_add_comment - Add comment to task

M365 Email

  • m365_read_emails - Read emails (limited to 25)
  • m365_send_email - Send email

GitHub

  • github_list_repos - List repositories
  • github_get_file - Get file contents (truncated if large)

Truncation Behavior

When responses exceed limits, the gateway:

  1. Truncates the data to fit within limits
  2. Adds _truncated: true to the response
  3. Includes _original_count and _returned_count
  4. Adds _message with pagination hints

Example truncated response:

{
  "success": true,
  "row_count": 100,
  "data": [...],
  "_truncated": true,
  "_original_count": 5000,
  "_returned_count": 100,
  "_message": "Results limited to 100 rows. Add LIMIT/OFFSET to query for pagination."
}

Version History

  • v1.0.0 - Initial release with response limiting
    • Based on SM-MCP-Gateway v1.8.0
    • Added ResponseLimits configuration
    • Added truncate_response utility
    • Snowflake, Asana, M365, GitHub tools

Support

For issues, contact the Sovereign Mind team or raise an issue in the repository.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选