eurlex-mcp-server

eurlex_structure

Read-onlyIdempotent

Returns an EU legal act's outline with chapter, article, annex, and judgment-paragraph offsets so you can jump directly to a specific section in the plain text.

Instructions

Returns the outline (table of contents) of an EU legal act — its chapters, sections, articles and annexes — each with a character offset into the document's plain text. Use it as a map for targeted reading: read an article's offset from the outline, then call eurlex_fetch(celex_id, format:"plain", offset, max_chars) with that offset to jump straight to that article instead of paging from the top of a long act. Identify the act by celex_id (e.g. "32024R1689"), eli, or oj_ref — provide exactly one. Each outline entry has: level (1=part/title/annex, 2=chapter, 3=section, 4=article), label (e.g. "Article 5", "CHAPTER III"), title (the heading's subtitle, e.g. "Prohibited AI practices"), and offset. total_headings is the full count; the returned list is capped at 300 for very large acts (truncated=true). Heading offsets are specific to the chosen language and to plain (tag-stripped) text — pass the SAME language to the follow-up eurlex_fetch call and keep format:"plain". Heading recognition covers English, German and French documents. For case-law documents (CELEX sector 6, e.g. CJEU judgments) the outline additionally lists each numbered judgment paragraph as "Paragraph N" (level 4); numbered-paragraph detection works in any language (it keys on the paragraph number, not heading words) — so you can get the offset of, say, paragraph 72 of a judgment and jump eurlex_fetch straight to it.

Input Schema

Name	Required	Description	Default
`eli`	No	European Legislation Identifier (ELI), short or full form, e.g. "reg/2016/679" or "http://data.europa.eu/eli/reg/2016/679/oj" (GDPR). Resolved to a CELEX ID via Cellar. Provide exactly one of celex_id, eli, or oj_ref.
`oj_ref`	No	Official Journal reference in the post-2023 scheme, e.g. "OJ:L_202401689" (AI Act). Resolved to a CELEX ID via Cellar. Provide exactly one of celex_id, eli, or oj_ref.
`celex_id`	No	CELEX identifier, e.g. "32024R1689" (AI Act). Provide exactly one of celex_id, eli, or oj_ref.
`language`	No	Language of the document to outline, as a Cellar 3-letter code (any of the 24 official EU languages, e.g. DEU, ENG, FRA). Heading labels are language-specific; the returned offsets index the plain text of THIS language, so pass the same language to the follow-up eurlex_fetch call.	ENG

Output Schema

Name	Required	Description
`note`	No	Present only when no headings were found or the outline was truncated
`outline`	Yes	Headings in document order
`celex_id`	Yes	The resolved CELEX ID (echoed for the follow-up eurlex_fetch)
`language`	Yes
`returned`	Yes	Number of headings in `outline` (<= total_headings)
`truncated`	Yes	True when `outline` was capped below total_headings
`source_url`	Yes
`total_chars`	Yes	Length of the plain text the offsets index into (matches eurlex_fetch total_chars)
`total_headings`	Yes	Total headings detected before the returned-list cap

Schema Changelog

Changes observed during successful MCP inspections.

Changed11 schema fields changedv2.4.4Oct 29, 2026
- removedInput schema / additionalProperties
```
Removed value: -false
```
- addedOutput schema / properties / outline / items / properties / level / maximum
```
Added value: +9007199254740991
```
- addedOutput schema / properties / outline / items / properties / level / minimum
```
Added value: +-9007199254740991
```
- addedOutput schema / properties / outline / items / properties / offset / maximum
```
Added value: +9007199254740991
```
- addedOutput schema / properties / outline / items / properties / offset / minimum
```
Added value: +-9007199254740991
```
- addedOutput schema / properties / returned / maximum
```
Added value: +9007199254740991
```
- addedOutput schema / properties / returned / minimum
```
Added value: +-9007199254740991
```
- addedOutput schema / properties / total_chars / maximum
```
Added value: +9007199254740991
```
- addedOutput schema / properties / total_chars / minimum
```
Added value: +-9007199254740991
```
- addedOutput schema / properties / total_headings / maximum
```
Added value: +9007199254740991
```
- addedOutput schema / properties / total_headings / minimum
```
Added value: +-9007199254740991
```

Changed2 schema fields changedv2.4.2Sep 7, 2026

removedInput schema / $schema

Removed value: -"http://json-schema.org/draft-07/schema#"

removedOutput schema / $schema

Removed value: -"http://json-schema.org/draft-07/schema#"

Changed1 schema field changedv2.3.0Jul 22, 2026
- changedInput schema / properties / language / default
```
Previous value: -"DEU"New value: +"ENG"
```
First observedv2.2.0Jul 10, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations already declare readOnly/idempotent/openWorld, and the description adds substantial behavior beyond them: a hard cap of 300 headings with a truncated=true flag, offsets being language- and format-specific, heading recognition limited to ENG/DEU/FRA, and paragraph-level outlining for CELEX sector 6 case-law that works in any language. These are exactly the non-obvious traits an agent needs to avoid misuse.

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?

Front-loaded with the purpose and the recommended call pattern before detail, and each sentence carries information (offsets, cap, language coupling, case-law paragraphs). It is on the long side and repeats some language-matching guidance already present in the schema, costing a point.

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

Completeness5/5

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

Despite an output schema existing, the description covers the return shape (level, label, title, offset, total_headings, truncated), the truncation edge case, cross-language offset pitfalls, and case-law paragraph exceptions. An agent has everything needed to call it and consume the result correctly.

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 100%, so baseline would be 3, but the description adds real semantics: heading recognition is only reliable for English, German and French documents, and offsets are tied to the language passed here, which must be reused on the follow-up fetch. Much of the 'provide exactly one of celex_id, eli, oj_ref' wording duplicates the schema, keeping this from a 5.

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?

Opens with a specific verb+resource: returns the outline (table of contents) of an EU legal act, enumerating chapters, sections, articles and annexes with character offsets. This is clearly distinguishable from siblings like eurlex_fetch and eurlex_search, and the tool's role as a navigation map is stated outright.

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?

Gives an explicit workflow: read an article's offset from the outline, then call eurlex_fetch(celex_id, format:"plain", offset, max_chars) to jump straight in 'instead of paging from the top of a long act'. It names the alternative tool and the precise condition that selects it, plus the constraint to reuse the same language and format.

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

Deploy Server