Skip to main content
Glama
matematicsolutions

mcp-nsa

by matematicsolutions
OverviewSchemaRelated ServersScoreDiscussions
JavaScript
Local

mcp-nsa

Installation (one command)

Published on npm + the MCP Registry (io.github.matematicsolutions/mcp-nsa). Run without cloning:

npx -y @matematicsolutions/mcp-nsa

MCP client configuration (stdio):

{ "mcpServers": { "mcp-nsa": { "command": "npx", "args": ["-y", "@matematicsolutions/mcp-nsa"] } } }

(Building from source - below.)

MCP License: MIT Node

MCP server for the case law of the Polish administrative courts - Naczelny Sad Administracyjny / NSA (Supreme Administrative Court) + 16 Wojewodzkie Sady Administracyjne / WSA (regional administrative courts) - via CBOSA (Centralna Baza Orzeczen Sadow Administracyjnych - the administrative courts' central case-law database) (orzeczenia.nsa.gov.pl).

Related MCP server: mcp-isap

Why

SAOS (the common-courts case-law analysis system) does not index the administrative courts. Yet that is exactly where the substantive case law lives:

  • GDPR / data protection - decisions of the President of UODO appealed to the WSA, cassation appeals to the NSA

  • Taxes - individual tax rulings, decisions of tax authorities

  • Customs and excise

  • Administrative permits, concessions

  • Review of public-administration activity

mcp-nsa closes this gap. Coverage: 2,390,000+ rulings (verified live 2026-07-08 across the full date range), from 1981 to today.

Tools

  • search(query, caseNumber?, court?, dateFrom?, dateTo?, pageNumber?)

    • search by keyword, case number, court, date. Fetches the top 5 full rulings (case number, court, date, panel, keywords, legal basis, excerpt).

  • get_judgment(doc_id) - full ruling by 10-character hex doc_id (from the CBOSA URL, e.g. 7E50984BB7).

  • search_by_case(caseNumber) - shortcut: search by case number (e.g. III OSK 1377/23, I SA/Gl 659/22).

Every response contains structuredContent.citations with the fields: title, url (CBOSA), case_number, court, judgment_date, decision_type, snippet, doc_id. Patron reads this field automatically and surfaces it in the UI panel as the section "Rulings from CBOSA (NSA / WSA - administrative courts)".

Stack

  • Node 18+

  • @modelcontextprotocol/sdk

  • Stdio transport

  • https + regex HTML parser (zero external deps beyond the SDK)

  • 500 ms throttle between requests (2 req/s)

  • SSL: rejectUnauthorized: false - the CBOSA chain is sometimes incomplete on some machines; public rulings, no PII, MITM risk negligible against the benefit. See: LDH issue #167.

Build + run

npm install
npm run build
node dist/index.js   # starts the server on stdio

Wiring into Patron

In patron/backend/mcp-servers.json (alongside mcp-saos and mcp-eu-sparql):

{
  "name": "nsa",
  "transport": "stdio",
  "command": "node",
  "args": ["C:/Users/<YOUR-USERNAME>/mcp-nsa/dist/index.js"],
  "enabled": true
}

Smoke test

echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"s","version":"0"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"search","arguments":{"query":"RODO art 6"}}}' \
  | node dist/index.js

Should return ~1500 hits + the top 5 with Polish NSA/WSA case numbers + CBOSA URLs

  • structuredContent.citations.

Lineage

Implementation rewritten in TypeScript on the basis of the HTTP/HTML contract from legal-data-hunter/sources/PL/NSA (Python + BeautifulSoup, MIT). It does not import the source code - it reproduces the query pattern and HTML parsing.

License

MIT.

Part of the MateMatic legal stack

This server is one of five MCP connectors covering Polish jurisdiction + EU law, used by Patron (AGPL-3.0) and any other MCP-aware legal AI agent.

  • mcp-nsa (this repo) - NSA + 16 WSA administrative courts (CBOSA)

  • mcp-saos - common courts, SN, TK, KIO

  • mcp-isap - Polish legislation (Dz.U. + M.P.)

  • mcp-krs - Polish company registry (KRS)

  • mcp-eu-sparql - EU law + CJEU (EUR-Lex)

All five MCP servers share the same structuredContent.citations contract: each tool returns an array of {title, url, snippet?, ...metadata} that legal agents can render directly in their citation panel.

See matematicsolutions/.github for the full org profile.

Available Tools

3 tools
get_judgmentA
Read-onlyIdempotent

Pobiera pelne orzeczenie z CBOSA po jego doc_id (10-znakowy hex). Zwraca metadane (sygnatura, sad, data, sklad, hasla tematyczne, podstawe prawna, typ orzeczenia) + pierwsze 2000 znakow tresci (sentencja + uzasadnienie). doc_id pochodzi z wynikow narzedzia 'search'.

ParametersJSON Schema
NameRequiredDescriptionDefault
doc_idYes10-znakowy hex doc_id z URL CBOSA, np. '7E50984BB7'.

TDQS

A4.4/5.0
Behavior4/5

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

Annotations declare readOnlyHint=true, idempotentHint=true, and description confirms it is a read operation. Adds behavioral detail that only first 2000 characters of content are returned.

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

Conciseness5/5

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

Two sentences, no wasted words. Action verb 'pobiera' front-loaded, structure clear.

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?

Without output schema, description adequately lists returned metadata fields and states content truncation at 2000 characters. Sufficient for agent to use tool.

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?

Single parameter doc_id with schema description providing format example. Description reinforces the format and source, adding value beyond schema.

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?

Description clearly states it retrieves a full judgment by doc_id, specifies the format (10-znakowy hex), and notes the doc_id comes from the 'search' tool, distinguishing it from siblings.

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

Usage Guidelines4/5

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

Explicitly says doc_id comes from 'search' results, implying usage after search. Does not explicitly state when not to use, but context of sibling tools provides differentiation.

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

search_by_caseA
Read-onlyIdempotent

Skrot: szuka orzeczenia po sygnaturze. Odpowiednik search z parametrem caseNumber. Jesli orzeczenie nie znajdzie sie - sygnatura moze byc z sadu powszechnego/SN/TK/KIO (uzyj wtedy saos__search_by_case).

ParametersJSON Schema
NameRequiredDescriptionDefault
caseNumberYesSygnatura akt, np. 'III OSK 1377/23'.

TDQS

A4.6/5.0
Behavior5/5

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

Annotations already declare readOnlyHint=true, destructiveHint=false, idempotentHint=true, openWorldHint=true. The description adds value by explaining this is a shortcut equivalent to search with caseNumber, and that it may not find judgments from certain courts, prompting a fallback. No contradiction with annotations.

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

Conciseness5/5

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

Very concise: two sentences convey purpose, equivalence, and fallback advice. Every sentence earns its place with no fluff.

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?

Given the tool's simplicity (one parameter, no output schema, safe annotations), the description is nearly complete. It states what it searches, the input format, and provides error handling guidance. Minor gap: no mention of the output structure, but it implies a judgment object. Almost complete.

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

Parameters3/5

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

Schema coverage is 100% with one parameter 'caseNumber' described with an example. The description adds that if the judgment is not found, the case number might be from other courts, but this is contextual rather than direct parameter semantics. Baseline 3 is appropriate.

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 description clearly states it searches for a judgment by case number ('szuka orzeczenia po sygnaturze'). It distinguishes from siblings by being a shortcut for 'search' with the caseNumber parameter and provides a fallback alternative (saos__search_by_case) for other court types.

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?

Explicitly tells when to use: as a shortcut for search with caseNumber. Also gives guidance on when not to: if judgment not found, case number might be from other courts (common, Supreme, etc.), then use saos__search_by_case. This clear when-to-use and alternative advice is excellent.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 3 tool updatesv1.1.0
    • First observedget_judgment
    • First observedsearch
    • First observedsearch_by_case
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.1/5.0
Disambiguation4/5

Tools are mostly distinct: search for general queries, search_by_case for case number lookup, and get_judgment for full retrieval by ID. The only minor overlap is that search_by_case is described as a shortcut for search with a case number, which could cause slight confusion but is clearly documented.

Naming Consistency5/5

All tool names use a consistent snake_case pattern with verbs (get, search) followed by nouns or qualifiers. The naming is uniform and predictable across the set.

Tool Count4/5

With 3 tools covering search, retrieval, and case-specific lookup, the set is minimal but reasonable for a focused legal database. The scope is narrow (administrative courts), so no heavy tool count is needed.

Completeness3/5

The tools cover basic search and retrieval, but notable gaps exist: lack of filtering by date, court level, or other metadata in search, and get_judgment only returns the first 2000 characters, limiting full-text access. Agents may need workarounds for deeper queries.

Maintenance

ActivitySlowing
ResponsivenessSyncing

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/matematicsolutions/mcp-nsa'

If you have feedback or need assistance with the MCP directory API, please join our Discord server