Repository avatar
Monitoring
v0.1.0
active

beamtable

io.github.CloudCorpRecords/beamtable

Search and compare a laser-materials dataset: columns, rows, stats, top values and provenance.

Documentation

beamtable

A table you can interrogate, reached over MCP. beamtable is a small forwarding client for the hosted Laser Materials Compare service: the tool list is served from your own machine using the captured definitions, and each call is sent upstream, with the service's answer returned exactly as it arrived.

Nine of the service's ten captured tools are exposed. You can learn the dataset's columns and provenance, search its rows, fetch rows by an exact value, set rows side by side for an "X vs Y" question, compute statistics for a numeric column, and rank rows by one. The tenth captured tool, submit_enquiry, is deliberately left out: it submits a person's enquiry answers — including the email address they give — to human providers who reply directly, and this rebuild does not send anyone's details anywhere. The two enquiry tools that remain, enquiry_describe and enquiry_fields, are published documentation reads about that flow; they take no personal details and submit nothing.

The tools

  • dataset_columns — The columns of the dataset, which of them are numeric, the row count and the provenance banner. No arguments; call it first to learn the schema.
  • dataset_search — Rows whose cells contain a query (case-insensitive), up to 50. An optional limit between 1 and 50 trims the list.
  • dataset_row — Rows where a named column equals a value exactly (case-insensitive).
  • dataset_compare — Rows whose column is any of 2 to 10 given values, in the order given — for side-by-side comparisons.
  • dataset_stats — Count, minimum, maximum, mean, median and sum of a numeric column; non-numeric rows are excluded and counted.
  • dataset_top — The highest rows by a numeric column, or the lowest when ascending is true, with an optional limit between 1 and 50.
  • dataset_provenance — The dataset's source, the date it was computed, its licence and how to cite it. No arguments.
  • enquiry_describe — What the enquiry flow does, who receives the details, the consent wording and how a person confirms. No arguments, and nothing is submitted.
  • enquiry_fields — Every field the enquiry asks: key, label, type, whether required, help text and allowed options. No arguments, and nothing is submitted.

Install and run

As an MCP server via npm:

npx -y @reneturcios/beamtable mcp

MCP client configuration:

{
  "mcpServers": {
    "beamtable": {
      "command": "npx",
      "args": ["-y", "@reneturcios/beamtable", "mcp"]
    }
  }
}

As a container:

docker run -i ghcr.io/cloudcorprecords/beamtable
docker run --rm ghcr.io/cloudcorprecords/beamtable dataset_columns '{}'

CLI

A single binary covers both the command line and the MCP server. Every tool is a subcommand whose one argument is a JSON string of that tool's inputs:

beamtable dataset_columns '{}'
beamtable dataset_search '{"query":"aluminium","limit":10}'
beamtable dataset_row '{"column":"material","value":"aluminium"}'
beamtable dataset_compare '{"column":"material","values":["aluminium","steel"]}'
beamtable dataset_stats '{"column":"melting_point"}'
beamtable dataset_top '{"column":"melting_point","limit":5}'
beamtable dataset_provenance '{}'
beamtable enquiry_describe '{}'
beamtable enquiry_fields '{}'
beamtable --version

Exit codes: 0 for a successful forwarded call, 1 when the result that comes back is an error, and 2 for usage mistakes like an unknown subcommand or a JSON argument that does not parse. submit_enquiry is not a subcommand and exits 2 like any other unknown name.

Configuration

  • LASERMATERIALS_MCP_URL — Replaces the upstream endpoint. Defaults to the pinned hosted endpoint, https://lasermaterialscompare.com/mcp. (This override variable is a reconstruction introduced by this rebuild.)

The Laser Materials Compare registry record lists no headers for its remote, so beamtable never attaches an authentication header — every call goes upstream keyless.

When the upstream cannot answer — it returns a JSON-RPC error, answers with a non-2xx status, sends a body that cannot be read, or the request itself fails — the caller receives an error result whose text begins with beamtable error:.

How forwarding works

The first call of a process opens a session against the upstream: an initialize request (protocol version 2025-03-26), a notifications/initialized notification, and then the tools/call. If the service issues an Mcp-Session-Id header during initialize, beamtable repeats it on subsequent requests and keeps reusing that one session until the process ends. Replies are understood in both plain JSON and server-sent-events framing. Successful results are passed through untouched, isError flag and any extra fields included.

Development

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 suite compares the served tool list against the captured surface tool for tool, asserts the omitted submit_enquiry tool is absent at the registry, MCP and CLI levels, walks the session handshake and its reuse, proves no Authorization header is sent under any environment, and confirms that arguments travel upstream unchanged and results return unchanged.

License

MIT — see LICENSE. Copyright (c) 2026 Rene Turcios.

NPM
@reneturcios/beamtable
Install Command
npm install @reneturcios/beamtable
Runtime: npx
OCI
ghcr.io/cloudcorprecords/beamtable:0.1.0
Install Command
docker pull ghcr.io/cloudcorprecords/beamtable:0.1.0:undefined
Runtime: docker