Lex: Luxembourg Law MCP Server
Server Details
Point-in-time access to Luxembourg law and ten EU acts: what any law said on a given date, not just the current text. 1,409 consolidated works and 4,705 dated versions from the official Legilux and EUR-Lex sources. Ten read-only tools: as-of text, timelines, per-article history, diffs between dates, and hash-verifiable provenance. No key.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-06-18
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 27 tools
The date/state tools (as_of, as_observed, status_on, in_force_on) and history tools (timeline, article_history, changes_in_period) have overlapping framing, though the descriptions distinguish them. Reference tools (citation, cited_by, relations) also require careful reading to choose correctly. The set is navigable with the descriptions, but misselection risk remains across several clusters.
All tool names use lowercase snake_case and are concise, but they mix verbs (ask, browse, resolve, search, verify), nouns (dossier, events, provenance), and prepositional query names (as_of, in_force_on). There is no predictable verb_noun pattern, though the names remain readable.
At 27 tools, the surface is well above the well-scoped 3-15 range and includes three refusal-only tools (concepts, knowable_on, transposition) that inflate the count. The complex legal domain justifies many retrieval modes, but the total still feels heavy.
Core read-only legal research paths are covered: browse/search/resolve, state snapshots, history, citations, provenance, verification, and evidence assembly. Explicit gaps remain for EuroVoc concepts, knowable-on reconstruction, and EU-Luxembourg transposition links, but these are transparently refused rather than silent dead ends.
Available Tools
27 toolsanswer_driftCInspect
The past dated answers a publisher revision invalidated, enumerated from the log's revising events; a genesis log holds none, and the answer says why rather than asserting that nothing drifted.
Inputs: after, limit, identifier.
article_historyCInspect
The history of one article of a Luxembourg act across its states.
Inputs: anchor (required), identifier (required), language.
askCInspect
The assistant, switched on as a labelled preview: the question, with obvious identifiers masked by code, is sent to the configured processor, which only plans which operations to run; code runs them, decides the verdict and writes every sentence from fixed wording; the answer is an answer_dossier (or the refusal advice_boundary), and a question it cannot answer gets assistant_v3_unavailable.
Inputs: question (required).
as_observedCInspect
The state of a Luxembourg act applicable on a date (publisher) as one build of the mounted log held it, named by that build's index digest, with its build time as an upper bound on observation; a time refuses, since no observation time is held.
Inputs: date (required), identifier (required), at, language, +1 more.
as_ofCInspect
The state of a Luxembourg act applicable on a date (publisher), with its articles; for an EU act, the wording dated at or before the date (an EU wording date, never an applicability date).
Inputs: date (required), identifier (required), language.
browseCInspect
The Luxembourg works this mount holds, one ordered and paged list with their identifiers, languages, state dates and the publisher's document types; filter by type or language.
Inputs: type, after, limit, language.
changes_in_periodDInspect
The Luxembourg states that began in a period, for a work or across the index.
Inputs: date_to (required), date_from (required), limit, language, +1 more.
citationDInspect
The references the publisher wrote in the text of a Luxembourg state's articles.
Inputs: date (required), identifier (required), after, limit, +2 more.
cited_byCInspect
The references, in any held Luxembourg state, whose target is exactly this work.
Inputs: identifier (required), after, limit.
classificationCInspect
The publisher's typed facts about a Luxembourg work, verbatim: document type, RDF types, legal value, responsible body, historical identifiers, publication and document dates.
Inputs: identifier (required), language.
conceptsAInspect
Not served: the EuroVoc descriptors, directory codes and subject matters of EU works, which no index holds; every call answers the typed refusal retrieval_mode_unavailable, naming the modes this server serves and the data that would serve it.
diffCInspect
The article-level difference between the states of a Luxembourg act on two dates.
Inputs: date_to (required), date_from (required), identifier (required), page, +1 more.
dossierCInspect
The work record of a Luxembourg act as the index holds it; for an EU work named by identifier, its held expressions from the EU index.
Inputs: identifier (required), language.
eventsCInspect
The Luxembourg event log, polled by cursor, at least once: a genesis log holds one first_sighting per held state; a chained log adds each later build's comparison events; each build's time, never an observation time.
Inputs: after, event, limit.
evidence_bundleCInspect
What a quote of a Luxembourg state needs, one bounded page at a time (at most 50 articles and 96 KiB of text; page or article chooses the page, contents adds the table of contents): each article's text with its digests, the state's permalink, digests, sources and official identities, composed under the rights rule; for an EU act, the same for the wording of the date.
Inputs: date (required), identifier (required), page, article, +2 more.
in_force_onDInspect
The Luxembourg works with a state applicable on a date.
Inputs: date (required), limit, language, identifier, +1 more.
knowable_onBInspect
Not served: what a reader could have known on a date, from publication and observation times, which no build records; every call answers the typed refusal retrieval_mode_unavailable, naming the modes this server serves (as_of, as_observed) and the data that would serve it.
manifestationDInspect
The publisher's manifestations of a Luxembourg work's expressions with their formats and items, the retained one marked with its body digest.
Inputs: identifier (required), format, language.
provenanceCInspect
The retained sources, rule profiles and digests behind a Luxembourg state.
Inputs: date (required), identifier (required), language.
relationsBInspect
The reference edges of a Luxembourg work in both directions, as one ordered, paged list.
Inputs: identifier (required), date, after, limit, +2 more.
resolveBInspect
Resolve a Luxembourg legal-act identifier or a hash-pinned permalink to the work and expression it names.
Inputs: identifier (required).
searchCInspect
With no work named, first read the query as an exact reference (an ELI, a CELEX, an EU act's name, a Memorial citation, a historical identifier, a permalink, or a citation of an article of an act at a date) and answer the address its act opens at, the acts it may name, or a typed refusal; otherwise search the held Luxembourg article text its rights admit (text they withhold is not matched), strict or relaxed, with the quotes that answer; for one EU work named by identifier, its held wording.
Inputs: query (required), language (required), date, mode, +3 more.
status_onCInspect
The publisher's force assertions about a Luxembourg work (its inForceStatus, dateEntryInForce and dateNoLongerInForce facts), verbatim, beside the state applicable on a date, with one fixed reading of the dates.
Inputs: date (required), identifier (required), language.
timelineBInspect
Every state of a Luxembourg act the index holds, in publisher applicability order; for an EU act, every wording its census discovered, in wording-date order.
Inputs: identifier (required), language.
transpositionAInspect
Not served: the transposition links between EU directives and Luxembourg measures, which are not acquired; every call answers the typed refusal retrieval_mode_unavailable, naming the modes this server serves (citation, cited_by, relations) and the data that would serve it.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceRemote MCP server that lets AI assistants read tracked competitive intelligence — competitor news, reviews, website changes, social posts and ads — and act on it by managing projects, competitors, battle cards, reports and canvases. Every answer carries a source and a date, and write actions only publish when explicitly shared.MIT
- AlicenseCqualityCmaintenanceCompetitor Monitor AI - MCP server providing AI-powered tools and automation by MEOK AI Labs112 npm37 PyPIMIT
- AlicenseAqualityDmaintenanceEnables brand visibility monitoring across major AI platforms like ChatGPT, Claude, Gemini, and Perplexity. It allows users to track visibility scores, analyze competitor data, and receive actionable insights to improve AI-generated brand recommendations.1611 npm1MIT

industrylens-mcpofficial
AlicenseNot gradedqualityCmaintenanceBrowse IndustryLens's published competitive-intelligence reports and head-to-head competitor comparisons from any AI agent — real, source-backed data.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.