datasetlas
README.md
# datasetlas
One dataset, seven lenses. datasetlas is a local forwarding client for the hosted DoorsetBook MCP service: it answers `tools/list` on your own machine from the captured tool definitions and forwards every query to DoorsetBook, returning the answers exactly as the service sends them back. Every answer cites the site, so figures stay attributable to their source.
Nothing is cached, stored or computed locally — the package is a pipe, not a copy of the data. Call `dataset_columns` first to learn the shape of the dataset, then search, look rows up, compare them, summarise a numeric column, rank by one, or read the provenance when you need to cite a number properly.
## The tools
- **`dataset_columns`** — The columns of the dataset, which of them are numeric, the row count and the provenance banner. No arguments; call this first to learn the schema.
- **`dataset_search`** — Rows whose cells contain the query, case-insensitive, up to 50. `query` is required; `limit` caps the count between 1 and 50.
- **`dataset_row`** — Rows where a column equals a value exactly, case-insensitive. Both `column` and `value` are required.
- **`dataset_compare`** — Rows whose column is any of the given values, in the order given, for "X vs Y" questions. `column` and `values` (2 to 10 values) are required.
- **`dataset_stats`** — Count, minimum, maximum, mean, median and sum of a numeric column. `column` is required; grouping commas and currency are handled upstream, and non-numeric rows are excluded and counted.
- **`dataset_top`** — The highest (or lowest) rows by a numeric column. `column` is required; `limit` and `ascending` adjust the ranking.
- **`dataset_provenance`** — The source, the date the data was computed, the licence and the citation for the dataset. No arguments; read this to attribute a figure correctly.
## Install and run
As an MCP server via npm:
```bash
npx -y @reneturcios/datasetlas mcp
```
MCP client configuration:
```json
{
"mcpServers": {
"datasetlas": {
"command": "npx",
"args": ["-y", "@reneturcios/datasetlas", "mcp"]
}
}
}
```
As a container:
```bash
docker run -i ghcr.io/cloudcorprecords/datasetlas
docker run --rm ghcr.io/cloudcorprecords/datasetlas dataset_search '{"query":"oak"}'
```
## CLI
One binary serves both the command line and the MCP server. Each tool is a subcommand taking a single JSON string with that tool's arguments:
```bash
datasetlas dataset_columns '{}'
datasetlas dataset_search '{"query":"oak"}'
datasetlas dataset_stats '{"column":"price"}'
datasetlas dataset_provenance '{}'
datasetlas --version
```
Exit codes: `0` when the forwarded call succeeds, `1` when the upstream result is an error, and `2` for usage problems such as an unknown subcommand or a malformed JSON argument.
## Configuration
- `DOORSETBOOK_MCP_URL` — Overrides the upstream endpoint. The default is the pinned hosted endpoint, `https://doorsetbook.com/mcp`. (The override variable is a reconstruction added by this rebuild.)
The upstream record declares no headers, so datasetlas sends no authentication header of any kind, under any environment. Upstream failures — a JSON-RPC error from the service, a non-2xx response, an unreadable body, or a request that fails outright — are reported to the caller as an error result whose text starts with `datasetlas error:`.
## How forwarding works
The first forwarded call opens a session: an `initialize` request (protocol version `2025-03-26`), followed by a `notifications/initialized` notification, then the `tools/call` itself. If the service answers the initialize request with an `Mcp-Session-Id` header, that value is carried on every later request, and the open session is reused for the rest of the process. Replies are accepted either as plain JSON or as server-sent events. Successful results pass through verbatim, including any `isError` flag and any extra fields the service includes.
## Development
```bash
npm install
npm run build # tsc -> dist/ (dist/cli.js is the bin)
npm test # tsx --test test/*.test.ts — stubbed upstream, no network
npm run dev # run the CLI from source
```
The test suite checks the served tool list against the captured surface, exercises the session handshake and its reuse, asserts that no `Authorization` header is ever sent, and verifies that arguments go upstream untouched and results come back untouched.
## License
MIT — see `LICENSE`. Copyright (c) 2026 Rene Turcios.
This server cannot be deployed
Maintenance
ActivityMaintained
ResponsivenessNo recent community issues