> For the complete documentation index, see [llms.txt](https://docs.glbnxt.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.glbnxt.com/platform/mcp-servers/dictum/method.md).

# Research method

The way an assistant should use DICTUM, as the server instructions and the `dictum-legal-research` skill describe it. The goal: answers a lawyer can rely on and check.

### The workflow

1. **Understand the question.** Area of law, articles, courts, periods, municipalities; is it about case law, the text of the law, or the legislator's intent?
2. **Search broadly.** `search_rulings` at least twice with different phrasings, at least one in Dutch legal terminology. Filters when the question implies them.
3. **Use the statute route.** `find_by_statute` for every article the question names or the rulings turn on: the article text first, then every ruling that cites it.
4. **Read the full text.** `get_ruling` for the three to five rulings that matter; never answer from search snippets alone.
5. **Check authority.** `citing_rulings` for the leading rulings: do later courts follow, refine or depart? Prefer the Hoge Raad and the other highest courts, and the Court of Justice for EU law; say when only lower-court rulings exist.
6. **Get the statutory text.** `get_article` for every article relied on, with `on=` the date that applied to the facts.
7. **Legislative history when intent matters.** `find_history` and `get_kamerstuk`.
8. **Trends when useful.** `facets` for how often and how courts decide; `coverage` when the answer depends on a source or period being present.
9. **Scholarly views when they help.** `search_literature` for open-access articles; cite them as an author's view, never in place of a ruling or an article of law.

### The memo

Written in the language of the question:

1. **Vraag**: the legal question in one or two sentences.
2. **Kort antwoord**: the conclusion in three to five sentences.
3. **Relevante rechtspraak**: per ruling the court, date, ECLI and link; what it decided in the assistant's own words; the decisive paragraph verbatim with its number.
4. **Wettelijk kader**: the articles, verbatim, with the version date if it matters.
5. **Wetsgeschiedenis**: only when it bears on the question.
6. **Analyse**: how the rulings fit together and apply to the question; the court's words kept apart from the assistant's reasoning.
7. **Onzekerheden**: gaps, conflicting rulings, missing coverage, facts that would change the outcome.
8. **Bronnen**: every ECLI, article and Kamerstuk used.

### The rules

* Quote freely and verbatim; the texts are public.
* Never invent an ECLI, date, paragraph number, article text or quote: everything cited comes from a tool result in the conversation.
* Cite to the paragraph: ECLI plus r.o. or punt; `art. 6:162 lid 2 BW`.
* Separate source from analysis.
* Say what a source is: a regulator's guidance, a dispute committee's ruling or a journal article is not binding law, and the result marks it "niet bindend".
* Say what DICTUM covers when a limit matters.
* One line of disclaimer at the end is enough: research support, the legal assessment stays with the user.

### Using it in your assistant

GLBNXT suggests a skill that carries this method, ready to add to Claude, Open WebUI, LibreChat or any other assistant and to adapt to your own practice: see [Suggested skill](https://docs.glbnxt.com/platform/mcp-servers/dictum/tools/skill).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.glbnxt.com/platform/mcp-servers/dictum/method.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
