mcp-data-service
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Logging level, e.g., DEBUG, INFO, WARNING, ERROR. Example: `LOG_LEVEL=DEBUG`. | |
| DATA_SOURCES | No | Comma-separated list of data sources in format `name=path`. For example: `default=./data/video_stats.csv,archive=./data/archive.db`. SQLite files are auto-detected by extension .db/.sqlite. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| extensions | {
"io.modelcontextprotocol/ui": {}
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| query_video_statsC | 查询统计数据。列名以实际 CSV 为准，metric 必须是存在的列。 |
| data_insightsC | 自动洞察：数据总览 + 异常下探（1.5σ 偏离）+ 建议。列名以实际 CSV 为准。 |
| list_columnsB | 列出数据集的全部列名与类型。 |
| list_sourcesA | 列出当前可用的数据源（通过 DATA_FILE / DATA_SOURCES 配置）。 |
| refresh_sourcesA | 清除数据源缓存；下次查询重新读取 CSV/SQLite。只读操作，不修改文件。 |
| get_schemaB | 获取数据集详细 Schema：列名、类型、非空数、均值/分位数、取值示例。 |
| preview_dataC | 分页预览数据，支持条件过滤。 |
| aggregate_statsC | 按字段分组聚合。 |
| correlation_analysisB | 数值列相关性矩阵（Pearson），可选指定列。 |
| sql_queryA | 只读 SQL 查询（仅 SELECT/WITH），在 SQLite 或 DataFrame 上执行。 |
| health_checkA | 健康自检：返回数据源、行列数、必要列、MCP 工具数。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| analyze_data | |
| anomaly_drilldown |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Most tools are distinct, but list_columns and get_schema clearly overlap, and query_video_stats, aggregate_stats, and data_insights have fuzzy boundaries around statistics and insights. Descriptions help somewhat, but an agent could easily select the wrong tool for a given analytical request.
All names are snake_case, but conventions are mixed: verb-noun names like list_columns and preview_data coexist with noun-phrase names like data_insights and correlation_analysis. query_video_stats also stands out as a domain-specific exception to the otherwise general-purpose naming pattern.
Eleven tools is an appropriate size for a data service covering source management, schema inspection, preview, aggregation, correlation, SQL querying, and health checks. The count is neither bloated nor too thin for the stated purpose.
The tool surface covers the core read-only data workflow well: discovering sources, inspecting schema, previewing, aggregating, running SQL, checking correlations, and getting automated insights. Minor gaps exist—such as no explicit dataset-comparison or export tool—but they are workable through sql_query and preview_data.