Agent Skills › Zafer-Liu/Data-Analysis-Agent › sql

sql

GitHub

提供安全只读SQL查询能力，先通过get_schema确认表结构，再使用query_data执行聚合或限流查询，禁止写操作，并解释结果。

skills/sql/SKILL.md Zafer-Liu/Data-Analysis-Agent

Trigger Scenarios

用户需要查询已授权数据表的记录用户请求基于特定条件的数据分析

Install

npx skills add Zafer-Liu/Data-Analysis-Agent --skill sql -g -y

More Options

Use without installing

npx skills use Zafer-Liu/Data-Analysis-Agent@sql

指定 Agent (Claude Code)

npx skills add Zafer-Liu/Data-Analysis-Agent --skill sql -a claude-code -g -y

安装 repo 全部 skill

npx skills add Zafer-Liu/Data-Analysis-Agent --all -g -y

预览 repo 内 skill

npx skills add Zafer-Liu/Data-Analysis-Agent --list

SKILL.md

Frontmatter

{
    "icon": "🗄️",
    "name": "sql",
    "description": "使用 SQL 安全查询当前已选择的数据表（query 数据查询）",
    "allowedTools": [
        "get_schema",
        "get_table_detail",
        "query_data"
    ]
}

SQL 查询

先读取已授权分析表的 schema，确认真实表名与字段名，再编写只读 SQL。优先聚合并限制返回行数，不猜测字段；SQL 数据源只能查询用户已选择的分析表。解释查询口径并总结结果。

Tool routing

Use get_schema before writing SQL to confirm real table and field names.
Use get_table_detail when a specific table needs deeper metadata.
Use query_data for read-only SQL. Prefer aggregate queries and explicit LIMIT for previews.
Do not use write, export, cleanup, or analysis tools from this skill.

Implementation reference

Schema/query tool entries: agent/tools/business/data.py::_tool_get_schema, agent/tools/business/data.py::_tool_get_table_detail, agent/tools/business/data.py::_tool_query_data
SQL validation: agent/validate.py