safe-data
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| schema_describeA | PII を除いたビュー(claude スキーマ)の列定義と、Faker で作ったダミー行を返す。実データは返さない。 |
| sql_runA | claude スキーマの PII 無しビューに対して SELECT を1本実行する(読み取り専用・行数上限・少数セル抑止)。 |
| files_describeA | PII inbox(Claude が直接読めない場所)にあるファイルの一覧と推定スキーマを返す。値は返さない。 ファイルは id で参照する(py_run の inputs に渡す)。 |
| fixture_makeA | ビュー名またはファイル id と同じ形の合成データ(Faker)を n 行返す。テストやスクリプト開発用。 |
| py_runA | analysis/ 配下のスクリプトを、ネットワーク無しのコンテナで inbox ファイルに対して実行し、小さな JSON 結果だけ返す。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool targets a distinct resource or operation (views, SQL, inbox files, synthetic fixtures, sandbox scripts). The main overlap is schema_describe and fixture_make, which both produce Faker dummy rows, though schema_describe is for schema inspection and fixture_make for generating test data.
All names use consistent snake_case with a clear object_action pattern: schema_describe, sql_run, files_describe, fixture_make, py_run. Minor abbreviations (sql, py) are readable and do not break the convention.
Five tools is well-scoped for a privacy-safe data access server. Each tool has a clear role in the workflow: schema discovery, querying, file cataloging, fixture generation, and sandbox execution.
The surface covers the core read-only analysis lifecycle: discover schemas, query safe views, inspect PII inbox metadata, generate synthetic data, and run sandboxed scripts. A minor gap is the lack of a tool to list available analysis scripts or otherwise help discover script paths for py_run.