biz-db-mcp
Enables agents to interact with MySQL databases securely, supporting multi-database read-only queries, protected DML writes, and streaming exports to Sandbox Datasets.
README
biz-db-mcp
一个面向 Agent 的 MySQL MCP Server,支持多数据库、只读查询、受保护的 DML 写入,以及批量导出到 Sandbox Dataset。
设计边界
- 默认只读:
BIZDB_WRITE_ENABLED默认关闭,单个数据库还必须显式allow_writes: true。 query与export只允许单条SELECT/WITH ... SELECT;execute只允许单条INSERT、UPDATE、DELETE或REPLACE。UPDATE/DELETE默认必须带WHERE,写入在事务中执行,并受最大影响行数限制。export使用服务端游标分批写 Parquet,再上传 Sandbox;身份从进程环境注入,模型不能指定目标 session。- SQL 审计只记录 SHA-256 和长度,不记录原始 SQL、参数或密码。
目录结构
biz_db_mcp/
config.py # 环境变量、多数据库、写入和 DBPM 配置
db.py # MySQL 连接、查询、写入
dbpm.py # DBPM 外部桥接命令适配器
sql_guard.py # SELECT/DML 语句边界
export.py # 流式 Parquet 导出
sandbox_client.py # Dataset 上传
audit.py # 结构化审计日志
server.py # MCP 工具注册
tests/ # 单元测试
配置
单库兼容配置:BIZDB_MYSQL_DSN=mysql://user:password@host:3306/database。
多库使用 BIZDB_DATABASES_JSON,格式可以是映射或 connections 数组:
{"connections":[
{"id":"primary","dsn":"mysql://user:password@db-a/app"},
{"id":"reporting","dsn":"mysql://user:password@db-b/reporting","allow_writes":false}
]}
BIZDB_DEFAULT_DATABASE 选择默认库。写入还需要 BIZDB_WRITE_ENABLED=true,并在目标库配置 allow_writes: true;可用 BIZDB_WRITE_MAX_AFFECTED_ROWS 和 BIZDB_WRITE_REQUIRE_WHERE 调整保护阈值。
工具包括 list_databases、describe_table、query、execute 和 export。execute 是单语句、事务提交的 DML 接口,不提供 DDL、事务控制或任意多语句执行。
DBPM(可选)
UPspec 中的 DBPM 是密码管理系统,不是 MCP 请求认证。Python 项目不内置私有 Java SDK;启用 BIZDB_DBPM_ENABLED=true 后,必须配置 BIZDB_DBPM_COMMAND、BIZDB_DBPM_SERVERS、BIZDB_DBPM_APP_NAME、BIZDB_DBPM_APP_KEY 和 BIZDB_DBPM_APP_IP。
桥接命令从 stdin 读取 JSON(包含 database、username、servers 等字段),只在 stdout 输出密码,非零退出码表示失败。DBPM 私钥通过 DBPM_APP_KEY 环境变量传给桥接进程,不出现在命令行参数中。数据库条目可用 dbpm_database/dbpm_user 覆盖 DBPM 查询键。
开发与运行
python3 -m pip install -e '.[dev]'
python3 -m pytest
biz-db-mcp
Sandbox 上传仍需配置 BIZDB_SANDBOX_BASE_URL、BIZDB_SANDBOX_API_TOKEN;导出身份需配置 BIZDB_SESSION_ID、BIZDB_ORG_ID、BIZDB_USER_ID。生产环境应使用 DBPM,不要把真实密码提交到配置、日志或测试夹具中。
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.