Skip to main content
Glama
openbot-coder

quantdb-mcp

by openbot-coder
OverviewSchemaRelated ServersScoreDiscussions
Python
Local

quantdb-mcp

量化金融数据库 + 只读 SQL MCP 服务。

  • 数据源：easy-tdx（通达信 TCP 行情）+ pywestockdata（腾讯/东财基本面 HTTP）

  • 存储：DuckDB 双文件 —— quantdb_ods.duckdb（ODS 原始层）/ quantdb.duckdb（DWD 层，清洗后）

  • 接口：FastMCP streamable-http 只读 query(sql) 工具（可跨库 JOIN）

分层架构

层

库文件

表

ODS

quantdb_ods.duckdb

stock_basic / daily_kline / realtime_quote / finance_income / finance_balance / finance_cashflow

DWD

quantdb.duckdb

dim_stock（去重标准化）/ fact_daily（列名统一）

APP

-

后续因子层（可复用 factor-pipeline 表达式引擎）

ODS 可整体重灌，DWD 由 sync_dwd 全量重建（幂等）。

Related MCP server: ashare-mcp

安装

cd projects/quantdb-mcp
uv sync                       # 或 pip install -e .

依赖：duckdb、fastmcp、pywestockdata、easy-tdx、click。

CLI

quantdb init                  # 建库建表（双文件）
quantdb full-load             # 全量同步：代码表 → 日K → 财务 → DWD 重建
quantdb full-load --codes 50 --kline-count 120   # 限前 50 只、每只 120 根K线（试跑用）
quantdb update --once         # 增量同步 + 一次实时快照
quantdb update                # 盘中 30s 快照循环（Ctrl+C 退出）
quantdb summary               # 各表行数与最新日期

数据目录通过环境变量 QUANTDB_DATA_DIR 指定（默认 ./data）。

MCP 接入

启动服务（streamable-http）：

uv run python -m quantdb.server.mcp   # 等价 pip 安装后的 quantdb-mcp

注册到 MCP 客户端（如 nanobot / Claude Desktop）：

{
  "mcpServers": {
    "quantdb": {
      "command": "quantdb-mcp",
      "type": "stdio"
    }
  }
}

查询示例（query 工具，只读）：

SELECT * FROM dim_stock WHERE market = 'sh' LIMIT 5
SELECT d.code, s.name, d.close FROM fact_daily d JOIN ods.stock_basic s ON d.code = s.code
  WHERE d.date = (SELECT max(date) FROM fact_daily)

安全限制

  • 仅允许单条 SELECT / WITH；写语句、多语句被拒绝

  • 表白名单：仅业务表（ODS 用 ods. 前缀），内部表不可见

  • 最多返回 10_000 行；查询超时 30s；DuckDB 以只读模式打开

测试

uv run pytest                    # 97 tests，100% 覆盖率（mock 数据源，不依赖网络）
uv run ruff check src tests      # 0 error

详见 design.md 与 task_list.md。

Available Tools

1 tool
queryA

对量化数据库执行只读 SQL 查询。

支持表：dim_stock、fact_daily、finance_income/balance/cashflow、 realtime_quote、stock_basic、daily_kline（ODS 表用 ods. 前缀）。 仅允许单条 SELECT/WITH；写语句与多语句会被拒绝；最多返回 10000 行。

ParametersJSON Schema
NameRequiredDescriptionDefault
sqlYes

Output Schema

ParametersJSON Schema
NameRequiredDescription

No output parameters

TDQS

A5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, but the description fully compensates by disclosing read-only behavior, statement restrictions, and the row limit. It also clarifies supported tables and the ODS prefix, which are important operational context. This exceeds typical descriptions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three terse sentences carry all key information: read-only, supported tables, and constraints. No filler or redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a single-parameter SQL tool with an output schema, the description covers the essential operational aspects: target database, allowed query forms, table names, and result limits. There are no obvious gaps that would prevent correct invocation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema only defines sql as a string with no description (0% coverage). The description adds crucial semantics: it defines the acceptable SQL dialect (single SELECT/WITH), enumerates permissible tables, and states the 10,000-row cap. This substantially compensates for the sparse schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb ('执行只读 SQL 查询' – execute read-only SQL queries) and identifies the resource ('量化数据库' – quantitative database), clearly distinguishing it as a query tool. It goes beyond a tautology by specifying supported tables and constraints.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It provides explicit usage boundaries: only single SELECT/WITH statements are allowed, write and multi-statement queries are rejected, and result size is limited to 10,000 rows. It also enumerates supported tables, giving the agent clear criteria for when this tool applies. No sibling tools exist, so alternative tool guidance is not applicable.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 1 tool updatev0.1.0
    • First observedquery
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.9/5.0
Disambiguation5/5

With only a single tool, there is no possibility of ambiguity or overlap between tools. The tool's purpose is clearly defined as executing read-only SQL queries against a specified set of tables, so agents can directly know what it does.

Naming Consistency5/5

The sole tool is named 'query', which is a simple, predictable, and consistent name. Since there is only one tool, there are no mixed conventions or inconsistent patterns to confuse an agent.

Tool Count4/5

A single tool might seem thin, but it serves a comprehensive purpose as a generic SQL query interface covering many tables. While additional metadata tools could be useful, the count is slightly under but acceptable for a focused read-only database server.

Completeness4/5

The tool provides complete read-only access to all listed tables, supporting SELECT/WITH queries with a row limit. A minor gap is the lack of schema/metadata exploration tools, but the description partially fills this by listing the available tables, making it workable for most use cases.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • ashare-mcp

    lolifamily
    A
    license
    A
    quality
    A
    maintenance
    A-share market data MCP server via baostock. Full-stack coverage: K-line, financials, DCF/DDM/PEG valuation, 11 technical indicators with proper split-day volume handling (OBV/MFI on raw bars), CN-style KDJ (J=3K-2D), risk metrics (Beta/Sharpe/MaxDD with stock-suspension-aware aligned returns), and PBoC macro data.
    25
    11
    MIT

  • stock-mcp

    jamespud
    A
    license
    B
    quality
    B
    maintenance
    MCP server that syncs and queries comprehensive stock data (quotes, financials, dividends, analyst forecasts, options, news) from Yahoo Finance and Investing.com into an external MySQL database.
    27
    200
    MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/openbot-coder/quantdb-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server