Belgium Data Desk

Draft a message to a dataset's publisher

contact_publisher

Read-only

Prepare a message to the people who publish a dataset, for the user to send themselves: the contact the publisher declared in its catalogue record (never a guessed address), a polite draft in the publisher's language (Dutch, French, English or German) naming the dataset, file, issue and the evidence checked, and a mailto: link. Nothing is sent by this tool. Use it when the user wants to report a problem (stale data, broken link, missing units or metadata, wrong values) or ask the publisher a question. Run get_dataset or preview_dataset first and pass what they showed as evidence; the answer says what to check before writing. Show the draft to the user to edit and send; send one message per problem, not one per related dataset. Example: {"id":"national:d04b704a-1672-4567-bf0d-9a9452618812","issue":"metadata","details":"The record says 2019 but the file has 2026 data."}.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "id": "<id>",
      "issue": "<issue>"
    },
    "name": "contact_publisher"
  }
}

Input Schema

Name	Required	Description
`id`	Yes	Dataset id from search_datasets or get_dataset.
`issue`	Yes	What the message is about: stale, access, metadata, units, values, missing, schema, licence or question.
`details`	No	What is wrong or what the user wants to know, in the user's words (at most 1,500 characters).
`evidence`	No	Up to 5 facts checked with the other tools, each with its source, e.g. 'preview_dataset: TIMESTAMP_EXTRACT goes up to 2026-11-01, catalogue says 2019-01-21'.
`language`	No	Language of the draft; default the publisher's (Dutch for Flemish bodies, French for Walloon and Brussels ones, else English).

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 15, 2026

TDQS

A4.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the readOnlyHint/destructiveHint=false annotations, it states 'Nothing is sent by this tool', that the address is never guessed, that the draft language follows the publisher, and that the response tells the agent what to check before writing. The only gap is that the exact shape of the returned draft/link is only loosely gestured at ('a mailto: link').

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The operative constraint ('nothing is sent', draft for the user) is front-loaded and every part carries information, including a useful worked example. It is dense with long compound sentences, so it is slightly harder to scan than it needs to be, but little is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a five-parameter, no-output-schema tool this covers prerequisites, evidence sourcing, language defaults, the no-send guarantee, and the user hand-off. The one soft spot is that the returned draft structure is only implied, so an agent must inspect the result rather than knowing its shape in advance.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is already 100%, so this is a bonus layer: the description tells the agent where the id comes from, that evidence should be facts harvested from get_dataset/preview_dataset with sources, and it supplies a concrete example object showing issue/details usage. The language default rule largely duplicates the schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence names a specific verb (prepare/draft a message) and a precise resource (the publisher contact declared in the catalogue record), and immediately scopes it as a draft the user sends themselves. This clearly separates it from any send or search sibling in the toolset.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit triggers are given (reporting stale data, broken links, missing units/metadata, wrong values, or asking a question), plus a hard prerequisite: run get_dataset or preview_dataset first and pass what they showed as evidence. It also states the one-message-per-problem rule and tells the agent to hand the draft to the user for editing rather than acting on it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Belgium Data Desk tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.