Italian Law Mcp

eu.ansvar由 ansvar.eu 发布v1.1.1更新于 Oct 25, 2026

Italian legislation via MCP — full-text search across statutes and provisions

发布方自己的服务:Ansvar: Dutch Law

概览

AI 生成的概览

本地 MCP 服务器,索引意大利法规,让助手检索法律、获取条文原文并核验引注。

可以做什么
按关键词检索意大利法律与法典,返回条文级结果,并可针对某个法律问题构建引注集合。获取某部法律中特定条文的全文,或不指定条号时获取全部条文。查询某部法律或条文是否现行有效、已被修订或废止,并校验或格式化法律引注。将意大利法律及具体条文对应到欧盟法律依据,查找某欧盟指令或条例在意大利的实施法律,并识别对已废止欧盟指令的引用。读取服务器元数据、数据集统计与数据来源许可信息。
适用场景
适合需要依据真实法条原文而非模型记忆来回答意大利法律或合规问题,以及需要把引注回溯到原始条文或对应欧盟指令与条例的场景。
运行要求
以本地进程方式运行 npm 包 @ansvar/italian-law-mcp,需要 Node.js 20 或更高版本,并为构建后的 SQLite 数据库预留约 500 MB 磁盘空间。数据库需通过仓库的采集脚本从上游来源构建;已发布的免费版快照仅部分收录。语料仅含意大利语。
安装前请注意
采集过程会下载、缓存并索引上游材料;README 要求使用者自行确认符合来源条款、署名要求、访问频率限制以及商业使用或再分发限制。免费版数据库为部分快照,未重新采集前会过时。README 还提到一个托管网关,需要 OAuth 注册且有付费层级,那是与本本地包不同的独立服务。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Italian Law Mcp,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

Italian Law MCP

The Italian law corpus is now served through the Ansvar Gateway. Connect your AI assistant (Claude, Copilot, Cursor, custom MCP client) to https://gateway.ansvar.eu/mcp — one OAuth connection, free tier available, covering this corpus plus EU regulations, national law across 28 audited jurisdictions, and CVE/security intelligence, every result with a verbatim source citation. Start at https://ansvar.eu/docs/quickstart

Connect

Claude Code (one line):

bash
claude mcp add ansvar --transport http https://gateway.ansvar.eu/mcp

Claude Desktop / Cursor — add to claude_desktop_config.json (or mcp.json):

json
{  "mcpServers": {    "ansvar": {      "type": "url",      "url": "https://gateway.ansvar.eu/mcp"    }  }}

Claude.ai — Settings → Connectors → Add custom connector → paste https://gateway.ansvar.eu/mcp

First request opens an OAuth signup flow (setup details: ansvar.eu/docs/quickstart). After signup, your client is bound to your account; tier (free / premium / team / company) determines fan-out, quota, and which downstream MCPs are reachable.


Self-host this MCP

You can also clone this repo and build the corpus yourself. The schema, fetcher, and tool implementations all live here. What is not in the repo is the pre-built database — TDM and standards-licensing constraints on the upstream sources mean we host the corpus on Ansvar infrastructure rather than redistribute it as a public artifact.

Build your own: run this repo's ingestion script (entry-point varies per repo — typically scripts/ingest.sh, npm run ingest, or make ingest; check the repo root).

MCP server for Italian law — statutes from www.normattiva.it with EU implementation mapping. Indexes the Codice Civile, Codice Penale, Codice Privacy (D.Lgs. 196/2003 + D.Lgs. 101/2018), the NIS2 transposition (D.Lgs. 138/2024), D.Lgs. 231/2001 corporate liability framework, and the Codice dell'Amministrazione Digitale.

[License: Apache 2.0] [MCP] [Jurisdiction]

What this is

This server indexes the legal materials listed under Sources below and exposes them via the Model Context Protocol. Part of the Ansvar MCP fleet — source-available servers published for self-hosting.

It makes no outbound network calls except to the upstream sources during ingestion — no analytics, no phone-home.

Coverage

  • Corpus target: 58,985 statutes, 267,838 provisions from Normattiva (the consolidated national legislative database)
  • Premium overlay: 759,592 preparatory works from the Italian Senate (Akoma Ntoso XML, CC-BY-4.0)
  • Jurisdiction code: IT
  • Languages: Italian only — no official translations

The corpus is rebuilt from the upstream sources by scripts/build-db.ts; re-run npm run build:db periodically to refresh. The current free-tier ship contains a partial-ingestion snapshot (~361 statutes); full re-ingestion is tracked as separate work.

See Sources below for source URLs, terms, and reuse conditions.

Why this exists

LLMs answering compliance, security, or legal questions from training data alone will fabricate citations — confidently producing article numbers, statute names, and source URLs that do not exist, or that do not say what the model claims. This MCP exists so an agent can call a tool that returns the real text, the real identifier, and the real source URL straight from the indexed materials — and ground an answer rather than recall it.

One MCP, one corpus. The point is composition.

The Ansvar Gateway (ansvar.eu) joins this MCP with the rest of the Ansvar fleet behind a single authenticated endpoint — 300+ servers covering legal jurisdictions, EU regulations, security frameworks, sector regulators, privacy-pattern catalogues, and risk-scoring tools. That lets an agent run cross-domain workflows that no single MCP can serve alone:

  • Threat model and TARA. Threat enumeration → known component vulnerabilities → severity scoring → applicable AI, cybersecurity, and automotive obligations → privacy threats. Every finding traceable to its source.
  • Gap analysis. Target framework requirements → current-state evidence → unmet obligations → remediation guidance and authority opinions. Every gap traceable to the specific requirement that flagged it.
  • Data Protection Impact Assessment. Privacy regulation articles → national DPA guidance (Garante per la protezione dei dati personali) → privacy-pattern catalogue → applicable case law.

Getting high-quality citations

Citation accuracy degrades when an agent's context fills up. Long inputs cause retrieval-stage drift — the model recalls claim text correctly but misattributes the source. Two practices keep accuracy high:

  1. Focused first pass, checking-agent second pass. Query a small, relevant set of MCPs first, then run a separate agent whose only job is to re-resolve each citation against the source MCP and flag any that no longer match. The checking agent uses the same MCP tools as the synthesis agent.
  2. Pull the source text verbatim when in doubt. Every citation an agent emits points back to a tool call against this server. You — or another agent — can call the same tool with the same identifier and read the raw statute, article, or standard text directly. If the verbatim text doesn't support what the agent claimed, the citation was misused, regardless of whether the identifier was real.

Both patterns work the same way self-hosted or through the gateway.

Tools

ToolDescription
aboutServer metadata, dataset statistics, freshness, and provenance
search_legislationSearch Italian laws and codes by keyword (in Italian); BM25 ranking, provision-level results
get_provisionRetrieve the full text of a specific article from an Italian law (or all articles if no article number)
list_sourcesReturns metadata about data sources backing this server (jurisdiction, authority, license)
validate_citationValidate an Italian legal citation against the database — does the cited law and provision exist?
build_legal_stanceBuild a citation set for a legal question by searching across all Italian legislation
format_citationFormat an Italian legal citation per standard conventions
check_currencyIs an Italian law or provision in force, amended, or repealed (abrogata)?
get_eu_basisGet EU legal basis (directives, regulations) for an Italian law
get_italian_implementationsFind Italian laws that implement a specific EU directive or regulation
search_eu_implementationsSearch for EU instruments that have been implemented or referenced by Italian laws
get_provision_eu_basisGet EU legal basis for a specific provision within an Italian law (article-level)
validate_eu_complianceCheck EU compliance status — detects references to repealed EU directives

Two ways to use it

Self-host (free, Apache 2.0) — clone this repo, run the build to compile the local SQLite database from the included data, point your MCP client at the local server. Instructions below.

Use the hosted gateway — for production use against the curated, kept-fresh corpus across the full Ansvar MCP fleet, with citation enrichment and multi-jurisdiction fan-out — see ansvar.eu.

Self-hosting

Requirements

  • Node.js 20 or newer
  • ~500 MB free disk for the built database (free tier)

Install

bash
git clone https://github.com/Ansvar-Systems/italian-law-mcp.gitcd italian-law-mcpnpm install

Build the database

bash
npm run buildnpm run build:db

build:db compiles data/source/ into data/database.db (the runtime SQLite the MCP queries). To refresh against the latest Normattiva snapshot, the ingestion script (scripts/ingest.ts) fetches from the upstream sources listed under Sources below; review the source's published terms before running ingestion in a commercial deployment.

Ingestion is a snapshot — your local copy goes stale until you re-run it. The hosted gateway corpus is refreshed continuously.

Configure your MCP client

json
{  "mcpServers": {    "italian-law-mcp": {      "command": "node",      "args": ["/abs/path/to/italian-law-mcp/dist/index.js"]    }  }}

For HTTP transport (Docker / Kubernetes), the runtime entry point is dist/http-server.js listening on PORT (default 3000) at /mcp.

Sources

SourceSource URLTerms / license URLLicense basisAttribution requiredCommercial useRedistribution / cachingNotes
Normattiva — consolidated Italian legislationhttps://www.normattiva.it/Note legaliPublic domain — Italian Copyright Law (Legge 633/1941) Art. 5 excludes official acts of the State from copyrightRecommendedConditionalConditionalItalian statutes themselves are not copyrighted; Normattiva's site terms restrict bulk reuse of the database structure. Verify use case.
Italian Senate — preparatory works (Akoma Ntoso XML)https://dati.senato.it/LicenseCC-BY-4.0RequiredYesYes759,592 preparatory works; included only in the premium overlay

Upstream license constraints

Ansvar attribution code: Italian-LDA-Art-5 (declared in infrastructure/attribution-licenses.json in the Ansvar architecture-documentation repo). The publisher (Normattiva) also declares CC-BY-4.0 as an overlay since 2026-01-01.

  • Normattiva. Italian official acts (statutes, decrees) fall outside copyright under Art. 5 Legge 633/1941 — atti ufficiali dello Stato e delle Amministrazioni pubbliche. Normattiva's published terms (note_legali.html) restrict bulk reuse of the database structure; reproducing the underlying legislative texts is permitted, but bulk re-publication of Normattiva's curated/consolidated database may not be. The included ingestion script extracts legislative text, not the database structure.
  • Italian Senate (premium overlay). Akoma Ntoso XML feeds at dati.senato.it are licensed CC-BY-4.0. Attribution is required; commercial reuse and redistribution are permitted. The premium overlay is not redistributed under this repository's Apache 2.0 license — see ansvar.eu for the hosted gateway corpus.

Coverage scope (narrow carve-out)

Italian Art. 5 is NARROW. Coverage is asymmetric across material types:

  • IN-SCOPE under Art. 5: statutes (leggi), decrees (decreti legislativi, decreti-legge), official acts of the State, decrees and decisions of public administrations in their administrative-act capacity, ministerial circulars.
  • OUT-OF-SCOPE under Art. 5: court judgments (sentenze) — Italian doctrine consistently treats court judgments as outside atti ufficiali for Art. 5 purposes. A court-decisions tier would need a separate basis (D.Lgs. 36/2006 PSI implementing the EU PSI Directive, or court-specific terms).
  • Why this matters: This MCP currently ships the statute tier (Normattiva) under Italian-LDA-Art-5. Case-law tier (Cassazione, Corte Costituzionale, giustizia amministrativa) is not in scope and would require a separate licensing declaration before ingestion.

See docs/audits/2026-05-17-eu-copyright-statutory-works-batch-1b-DE-IE-IT-NL-ES.md §3 in the Ansvar architecture-documentation repo for the verbatim Art. 5 text and the coverage analysis.

What this repository does not provide

This repository's source — the MCP server code, schema, and ingestion script — is licensed under Apache 2.0. The license below covers the code in this repository only; it does not extend to the upstream legal materials.

Running ingestion may download, cache, transform, and index materials from the listed upstream sources. You are responsible for confirming that your use of those materials complies with the source terms, attribution requirements, robots/rate limits, database rights, copyright rules, and any commercial-use or redistribution limits that apply in your jurisdiction.

License

Apache 2.0 — see LICENSE. Commercial use, modification, and redistribution of the source code in this repository are permitted under that license. The license does not extend to upstream legal materials downloaded by the ingestion script; those remain governed by the source jurisdictions' own publishing terms (see Sources above).

The Ansvar gateway

If you'd rather not self-host, ansvar.eu provides this MCP plus the full Ansvar fleet through a single authenticated endpoint, with the curated production corpus, multi-MCP query orchestration, and citation enrichment.


Issues: github.com/Ansvar-Systems/italian-law-mcp/issues · Security: [email protected]

来源:README.md,提交 6b5b87f

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v1.1.1最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

其他 Ansvar 集成