Borrowing it
Nothing to install: this file belongs to redgreat/db_mcp_server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/redgreat/db_mcp_server/main/AGENTS.mdgit clone --depth 1 https://github.com/redgreat/db_mcp_serverWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/redgreat/db_mcp_server/agents-md)<a href="https://agentmods.dev/instructions/redgreat/db_mcp_server/agents-md"><img src="https://agentmods.dev/badge/instructions/redgreat/db_mcp_server/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/redgreat/db_mcp_server/agents-md"><img src="https://agentmods.dev/badge/instructions/redgreat/db_mcp_server/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03455 | $0.03455 |
| Opus 5.5 | $0.01382 | $0.01382 |
| Sonnet 5.5 | $0.00691 | $0.00691 |
| Haiku 4.5 | $0.00346 | $0.00346 |
Grade A, and why
db_mcp_server AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — DB MCP Server
本文件是 AI 编码智能体迭代本仓库的入口指引。先读这里,再按引用深入具体文件。 面向「工具调用方」的规范见
.ide_mcp_rules.md/.cursor_mcp_rules.md,本文件面向「代码修改方」。
1. 项目是什么
企业级数据库访问代理(MCP Server)。客户端(AI IDE、HTTP 调用方)通过访问密钥 + 连接级授权访问 MySQL/PostgreSQL,不直接接触真实数据库凭证。凭证加密存放在独立的 PostgreSQL 管理库中。
核心能力:
- 标准 MCP 协议(SSE + JSON-RPC)+ 简化 HTTP API,同一 FastAPI 进程、同一端口
- 安全:SQL 风险拦截、连接级权限(只读/写/DDL)、IP 白名单、数据脱敏、审计日志
- 20 个 MCP 工具:元数据浏览、查询执行、文档/ER/数据流导出、SQL 审查、性能分析、Schema 对比、Mock 数据、报表上传 OSS 等
- 可选 LLM 增强(
src/ai/):SQL 审查、调优建议、ER 业务解读 - SvelteKit 管理后台(构建产物在
src/static/,与后端同源同端口)
2. 技术栈与运行
| 层 | 技术 |
|---|---|
| 后端 | Python 3.13 · FastAPI · SQLAlchemy(Core 风格,Table+autoload)· Uvicorn |
| 管理库 | PostgreSQL(表结构由 src/admin/models.py: ensure_schema() 用 meta.create_all 管理,无迁移工具) |
| 前端 | SvelteKit 2 · Svelte 5(runes)· TypeScript · Tailwind 4 + DaisyUI · adapter-static |
| 部署 | Docker(python:3.13-slim,supervisord,非 root)· GitHub Actions 打 v* tag 推阿里云 ACR |
常用命令:
# 后端(项目根目录,配置文件必需)
cp config/config.yml.example config/config.yml # 改 master_key / admin_database
pip install -r requirements.txt
python src/init_admin_db.py # 初始化管理库(建表+默认admin/admin123)
uvicorn src.server:app --reload --host 0.0.0.0 --port 3000
# 前端(构建产物输出到 ../src/static/)
cd frontend && npm install && npm run build # 改前端必须重新 build
cd frontend && npm run check # svelte-check 类型检查
# 验证(Makefile;Windows 下用 git-bash 跑 make,或直接跑下面的命令)
pytest test/ # Python 测试(目前仅占位+sql_guard)
flake8 src/ test/ # lint(仓库历史代码有既有告警,只保证不新增)
cd frontend && npm run check
# MCP 工具联调(需服务已启动且有有效密钥)
python script/mcp_call.py --tool list_connections
配置文件通过 CONFIG_PATH 环境变量可覆盖,默认 config/config.yml(相对项目根目录,src/config.py 用 dataclass 强校验加载)。
3. 目录结构(改动前先定位层)
src/
├── server.py # FastAPI 应用工厂:静态挂载、/query、/sse/query、/metadata/*、/transaction/*、SPA 兜底
├── config.py # YAML → dataclass 配置(新增配置项要改这里 + config.yml.example)
├── init_admin_db.py # 管理库初始化脚本入口
├── logging_utils.py # get_logger(name, log_dir):滚动文件日志,日志时间走 timezone_util
├── timezone_util.py # 时区序列化/日志时间
├── mcp/
│ ├── tools.py # ★ 工具 schema 注册表 MCP_TOOLS(改工具定义改这里)
│ ├── server.py # HTTP 协议路由 /mcp/tools、/mcp/call → _execute_tool 分发
│ ├── standard_protocol.py # ★ SSE 协议路由 /mcp/sse、/mcp/message → _execute_mcp_tool_inner 分发
│ └── permissions.py # MCP 侧访问密钥/连接权限校验
├── security/ # sql_guard(权限分级) / sql_analyzer+interceptor(风险评分) /
│ # ip_whitelist / data_masker / secret(Fernet加解密) / client_ip
├── tools/ # 20 个工具的业务实现,db_*_tool.py 命名;纯函数风格,接收 engine 参数
├── db/db_operations.py # QueryProxy:engine 缓存池 + 事务管理(txn_map + 清理线程)
├── admin/
│ ├── web.py # /admin/* 路由聚合入口(include 各子路由)
│ ├── context.py # ★ AdminContext:engine/tbl/auth/日志器等共享依赖 + 测试连接 helpers
│ ├── schemas.py # pydantic 请求模型
│ ├── routes/ # ★ 子路由按资源拆分:auth/users/keys/connections/permissions/
│ │ # whitelist/llm/logs/misc,各导出 build_router(ctx)
│ ├── models.py # ensure_schema() 定义管理库全部表 + 默认数据初始化
│ ├── auth.py # JWT + bcrypt 认证
│ └── schema_cache.py # 管理库 Table 反射缓存(autoload_with 只反射一次)
├── ai/ # llm_client(读库取激活配置) / service(llm_ask 统一入口) / context / llm_active
├── logging/ # audit_logger(写库) / llm_call_logger / system_logger
├── services/oss_sts_service.py # STS 临时凭证上传
└── static/ # 前端 build 产物,勿手改
frontend/src/routes/ # (app)/ 下是受保护页面:connections/keys/audit/llm-logs/ai-settings/users;login/ 独立
frontend/src/lib/api.ts # 同源 fetch 封装,Bearer token,401 自动跳登录
config/config.yml.example # 配置模板,新增配置项必须同步更新
script/ # mcp_call.py(SSE测试) / redeploy.sh / release.sh / dockerbuild.ps1 / reset_admin_password.py
docs/ # 面向使用者的中文文档
test/ # pytest,测试基建很薄,新增逻辑请补测试
.github/workflows/ # build-deploy.yml:push v* tag → buildx → 阿里云 ACR
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 171 lines · 3,455 tokens per session scan A 7f421ccebaa1
db_mcp_server AGENTS.md is an instructions file published in the GitHub repository redgreat/db_mcp_server (2 stars, last pushed 11d ago), licensed MIT. It adds 3,455 tokens to every session, about $0.0138 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-27.
db_mcp_server AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| db_mcp_server AGENTS.mdredgreat/db_mcp_server | Instructions file | 2 | 11d ago | MIT | A | 3,455 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |