Skip to main content
Glama

Corpus Iuris — Mexican law for AI agents

Mexican law for AI agents. Search 6,710 federal and state statutes, 311,701 Supreme Court tesis and 11,708 precedents — with verifiable citations, dated validity and a public URL for every document.

MCP Registry API OpenAPI


The problem

AIs invent Mexican law. They cite tesis with registration numbers that don't exist, articles that were repealed, and jurisprudence that is actually an isolated tesis. In litigation, that costs a case.

Related MCP server: saij-mcp

The solution

A Mexican legal corpus queryable via API or MCP, where every result carries the actual text, its dated validity and a citable public URL.


Connect in one click (MCP)

Remote MCP server with OAuth 2.0 (dynamic registration RFC 7591 + PKCE S256):

https://api.corpusiuris.mx/mcp

Claude Desktop / Claude.ai → Settings → Connectors → Add custom connector → paste the URL. OAuth is discovered automatically.

Any MCP clientclaude_desktop_config.json, Cursor, Windsurf, VS Code:

{
  "mcpServers": {
    "corpus-iuris": {
      "url": "https://api.corpusiuris.mx/mcp"
    }
  }
}

Available tool: buscar_derecho_mexicano(query, fuentes).


No registration: the REST API

No API key. No account. Copy and paste:

curl "https://corpusiuris.mx/api/agent/v1/search?q=requisitos+de+la+orden+de+cateo&fuentes=leyes,tesis"

The three routes

Route

What it's for

GET /api/agent/v1/search?q=…&fuentes=…

Search. Up to 5 results per source.

GET /api/agent/v1/documento?tipo=ley&id=slug/num

Full text. Doesn't use quota.

GET /api/agent/v1/stats

How many documents there are, up to date.

Full contract: openapi.json (OpenAPI 3.1).


What makes a response citable

{
  "tipo": "ley",
  "referencia": "Art. 17",
  "titulo": "LEY DE AMPARO, REGLAMENTARIA DE LOS ARTÍCULOS 103 Y 107…",
  "url": "https://corpusiuris.mx/ley/ley-de-amparo…/articulo/17",
  "vigencia": {
    "estado": "VIGENTE",
    "fecha_ultima_reforma": "2025-10-16",
    "verificado_al": "2026-08-17",
    "fuente_oficial": "https://www.diputados.gob.mx/LeyesBiblio/pdf/LAmp.pdf",
    "como_citar": "Di 'vigente al 2026-08-17' y enlaza la fuente oficial."
  }
}

And in case law, the data point that avoids the classic mistake:

"fuerza": {
  "es_jurisprudencia_obligatoria": false,
  "como_citar": "Tesis AISLADA: orientadora, NO obligatoria. No la cites como jurisprudencia."
}
  • Lawsvigencia, with the date a night watchman verified the text against the official source. A repealed law comes out with ADVERTENCIA.

  • Tesisfuerza: binding or isolated.

  • Precedentsfuerza: they are rulings, not jurisprudence by themselves.


Quotas

Lane

Quota

How it's recognized

AI platform

3,000/day per platform

IP verified against the official list it publishes

Anonymous agent

500 per IP + 10/day forever

IP only

Free account

500 + 20/day for life

ci_agent_ token or MCP

Paid

3,000/month · unlimited

token

The anonymous lane's daily floor never runs out: an IP never stays dead permanently.

IP-verified platforms: OpenAI, Anthropic, Google, Perplexity, Meta. The User-Agent header isn't enough — it's checked against the official feed.


Two rules

  1. Cite the source. «Corpus Iuris (corpusiuris.mx)» + the url link of each result you use. Legal texts are public information; the search and curation belong to Corpus Iuris.

  2. Don't cache. The corpus changes every day: the watchman scans ~1,000 laws each night, the DOF publishes reforms and the SCJN uploads tesis. Responses are served with Cache-Control: no-store for that reason.


Examples


In English

Corpus Iuris is a Mexican legal corpus built for AI agents. Free REST endpoint (no key), native MCP server with one-click OAuth, and a public citable page for every document. Every result carries dated validity and, for case law, whether it is binding jurisprudence or a merely persuasive tesis aislada — the distinction that most often trips up AI in Mexican law.

Docs: https://corpusiuris.mx/agentes · OpenAPI: https://corpusiuris.mx/openapi.json


Does not replace the official source

It is a research tool of a referential nature. For a formal filing, verify in the DOF, the Semanario Judicial or the state's official gazette — every result carries the direct link.

Operated by AteraSoft (Mexico) · contacto@corpusiuris.mx

F
license - not found
Not graded
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

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

Related MCP Servers

  • A
    license
    Not graded
    quality
    Not graded
    maintenance
    Provides access to Argentina's official legal database (SAIJ) to search and retrieve judgments, legislation, summaries, and legal doctrine. It allows AI clients to query Argentine legal information and access metadata for various legal documents.
    6
  • A
    license
    A
    quality
    D
    maintenance
    Connects AI clients to Argentina's official legal database (SAIJ) to search and retrieve court rulings, legislation, summaries, and legal doctrine. It enables users to perform keyword searches and fetch detailed metadata and summaries for Argentinian legal research.
    3
    2
    MIT
  • A
    license
    Not graded
    quality
    B
    maintenance
    Provides AI assistants with live, authenticated access to official Mexican data sources including CURP, RFC, postal codes, phone numbers, SPEI payments, CFDI status, DOF semantic search, and geocoding via INEGI/INE.
    MIT

View all related MCP servers

Related MCP Connectors

View all MCP Connectors

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/maharbanaretsagrav/corpus-iuris-mcp'

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