Leyes Como Código - Mexico

[!IMPORTANT] Tezca handles legal/compliance data, official-source scraping, search indexes, API keys, webhooks, chat/LLM prompts, billing handoffs, and package/deploy publishing paths. Scrapers, RMF/DOF ingestion, index rebuilds, exports, webhooks, Selva chat, DB mutations, and publish/deploy flows are side-effectful. Keep examples placeholder-only and set the matching local guard env var only for an explicit operation.

The definitive digital platform for Mexican legal research - comprehensive, machine-readable database of Mexican laws (federal, state, municipal) with intuitive interfaces for professionals and citizens.

Coverage: 93.9% of Legislative Laws (11,696 of 12,456) + 18,439 non-legislative (30,343 total) — [sourced from data/universe_registry.json] Status: Production Ready

Quick Start

Prerequisites

• Node.js 20+
• Python 3.10+
• Docker & Docker Compose

Development Setup

1. Install Dependencies

   npm install          # Frontend (all workspaces)
   poetry install       # Backend (Python)
   cp .env.example .env # Configure environment
   

2. Start Development Servers

   npm run dev:web          # Public portal → http://localhost:3000
   npm run dev:admin        # Admin console → http://localhost:3001
   poetry run python manage.py runserver  # API → http://localhost:8000
   

3. Docker (all services)

   docker compose up -d   # API, Celery, Web, Admin, PostgreSQL, Redis, Elasticsearch
   

Coverage

All numbers sourced from data/universe_registry.json with links to official sources.

Level	Laws	Universe	Coverage	Source
Federal	333/336	336	99.1%	Cámara de Diputados
State (Legislativo)	11,363/12,120	12,120	93.7%	OJN - Poder Legislativo
State (Other Powers)	18,439/23,660	23,660	77.9%	OJN - Poderes 1/3/4
Municipal	208	Unknown	N/A	5 city portals (no census exists)
Leyes Vigentes	11,696/12,456	12,456	93.9%	Federal + State Legislativo

Note: 782 OJN links are permanently dead (Michoacán 504, EDOMEX 141, SLP 47). Municipal universe is unknown — INEGI reports 2,468 municipalities but no authoritative count of municipal laws exists.

Features

• ✅ 93.9% Legislative Coverage - 11,696 of 12,456 legislative laws (federal + state)
• ✅ Structured Parsing - Akoma Ntoso XML output with automated validation
• ✅ Dynamic Dashboard - Real-time statistics and recent legislation feed
• ✅ Advanced Search - Date range filtering, state filters, and auto-complete
• ✅ Law Detail 2.0 - Enhanced typography, improved ease-of-reading, and citations
• ✅ Quality Validation - 5 automated checks, A-F grading
• ✅ Full-Text Search - 1,100,000+ articles indexed in Elasticsearch
• ✅ Version History - Track legal evolution over time
• ✅ REST API - Machine-readable access for legal tech (paginated, filtered, rate-limited)
• ✅ Batch Processing - Parallel ingestion with 4-8 workers
• ✅ Production Ready - Full-stack testing (2164 Pytest + 930 web Vitest + 78 admin Vitest + 48 api-client Vitest + 23 MCP server Pytest); backend coverage gate 60% (≈64% actual), frontend gates 61% statements / 62% lines with all: true denominator
• ✅ OpenAPI Documentation - Swagger UI, ReDoc at /api/docs/
• ✅ Background Processing - Celery + Redis for ingestion jobs
• ✅ Cross-References - Automatic detection and linking between laws
• ✅ Legal Pages - Terms & Conditions, Legal Disclaimer, Privacy Policy (trilingual ES/EN/NAH)
• ✅ Site Footer - Persistent navigation, official source links, disclaimer bar
• ✅ Disclaimer Banner - Dismissable one-time homepage notice (localStorage persistence)
• ✅ Trilingual UI - ES/EN/NAH (Classical Nahuatl) language toggle across all components (law content remains Spanish)
• ✅ Tezca Manifesto - /acerca-de brand page with mission statement
• ✅ Persistent Navbar - Sticky navigation with mobile hamburger menu, transparent-on-homepage
• ✅ Bookmarks - Heart toggle, localStorage persistence, /favoritos page
• ✅ Reading UX - Progress bar, font size control, back-to-top, breadcrumbs
• ✅ Share & Export - Social sharing (Twitter, LinkedIn, WhatsApp), copy link, PDF print export
• ✅ - Shaped placeholders for law detail, search results, dashboard

Architecture

Monorepo Structure

This project uses a monorepo architecture managed by NPM Workspaces.

/
├── packages/
│   ├── ui/          # Shared UI Library (@tezca/ui) - React 19 / Shadcn
│   ├── lib/         # Shared Utilities & Types (@tezca/lib)
│   ├── api-client/  # TypeScript SDK (@tezca/api-client)
│   └── tsconfig/    # Shared TypeScript configurations
├── apps/
│   ├── web/         # Public Portal (Next.js 16)
│   ├── admin/       # Management Console (Next.js 16)
│   └── api/         # Backend API (Django / Python)
└── package.json     # Workspace Root

Components

• Ingestion Pipeline: PDF Download → Text Extraction → Parsing → Validation
• Public Portal: Citizen-facing search and traversal of laws
• Admin Console: Operator dashboard for monitoring ingestion jobs

Role in the MADFAM monetization engine

Tezca is intended as a revenue-carrying product in MADFAM's compliance line, not an internal tool. It is aimed at the people whose work depends on knowing what Mexican law currently says — compliance officers, accountants, legal and regulatory teams — with the machine-readable API over the legal corpus as the principal commercial surface. Tezca is also an upstream data dependency for other MADFAM compliance products, which consume its feeds rather than re-deriving them.

Its place in the wider MADFAM pipeline is simply that of a product to be sold. Demand discovery, pricing, checkout, invoicing, and entitlement are each owned by other platforms; none of that path lives in this repository, and nothing here should grow a dependency on it.

Boundary note. This is a deliberately sanitized statement of Tezca's designed role. It is not a statement about the live state of any deployment, nor about current commercial availability. The canonical end-to-end description of the monetization pipeline is private and lives in the internal-devops repository at docs/monetization-engine.md. Catalog prices, payment and entitlement wiring, credential names, deployment topology, and tax/invoicing configuration belong there and must not be added to this repository.

Last Updated: 2026-07-26

Documentation

• Setup Guide - Installation and configuration
• Tech Stack - Approved technologies
• Architecture - System design
• Testing - Test suite (backend + frontend). How to run each suite, the skip inventory, known flaky tests and the manual PDF render check: docs/guides/TESTING_STRATEGY.md
• MCP server - tezca-mcp: tools, /health, host protection, Docker
• Security policy and baseline - Dependency security floors and how to ship a root-lockfile fix
• Production deployment - Deploy workflows, trigger paths, verification
• llms.txt - Agent-consumable project summary (llms-full.txt for expanded version)

Related repositories / contracts

Repository	Contract	Defined in
janua	Identity provider. apps/web and apps/admin are Janua OIDC clients; the API validates Janua RS256 JWTs (see docs/AUTH.md).	docs/guides/ECOSYSTEM_INTEGRATION.md
janua	Machine-to-machine tokens, for a service that calls Tezca's API with a Janua token instead of an API key.	docs/service-tokens.md
enclii	Deploy platform. enclii.yaml follows its service spec; the deploy-*.yml workflows build, pin digests in k8s/production/ and report to Enclii, which ArgoCD then syncs.	docs/reference/service-spec.md, docs/guides/EXTERNAL_REPO_DEPLOY.md
karafiel (consumer)	Reads Tezca's law feeds: the fiscal values feed and the labor rules (SAT catalogues) instead of hard-coding them.	This repo: docs/FISCAL_VALUES_FEED.md,

Performance

Metric	Result
Processing Speed	23s per law
Parallel Speedup	3-4x
Schema Compliance	100%

Project Roadmap

Phase 1: Federal Laws - ✅ COMPLETE

• ✅ 333 federal laws ingested (99.1% of 336)
• ✅ Quality validation framework
• ✅ Elasticsearch full-text search

Phase 2: State Laws - ✅ COMPLETE

• ✅ 11,363 state laws downloaded (93.7% of 12,120 OJN Legislativo)
• ✅ Database schema update
• ✅ State law processing pipeline
• ✅ Frontend state filters

Phase 3: UI/UX Transformation - ✅ COMPLETE

• ✅ Dynamic Homepage Dashboard
• ✅ Law Detail Page 2.0 (breadcrumbs, font control, progress bar)
• ✅ Advanced Search with Autocomplete Typeahead
• ✅ Legal Pages (Terms, Disclaimer, Privacy) — trilingual ES/EN/NAH
• ✅ Persistent Navbar + Site Footer + Disclaimer Banner
• ✅ Comparison Tool (side-by-side, sync scroll, mobile tabs)
• ✅ Bookmarks, Share Buttons, PDF Export, Loading Skeletons
• ✅ API Hardening (pagination, filtering, rate limiting, search-within-law)
• ✅ CI/CD (coverage, E2E in CI, Dockerfile verification)

Phase 4-7: Data Depth & Infrastructure - ✅ COMPLETE

• ✅ 150 federal reglamentos ingested
• ✅ 18,439 non-legislative state laws (77.9% of OJN Poderes 1/3/4)
• ✅ Parser V2 (TRANSITORIOS boundary, dedup, Bis patterns)
• ✅ DOF daily scraper wired to Celery Beat (7 AM)
• ✅ Dual storage backend (local / Cloudflare R2)
• ✅ Sentry integration, ES resilience (retry/timeout/pooling)
• ✅ JSON-LD structured data on law detail pages
• ✅ Trilingual UI (ES/EN/NAH — Classical Nahuatl)

Phase 8-9: Surface & Search Intelligence - ✅ COMPLETE

• ✅ law_type field + non-legislative badges
• ✅ Faceted search (ES aggregations), browse by category/state
• ✅ Related laws (more_like_this), sort on catalog
• ✅ Spanish URL paths (/leyes, /busqueda, /comparar) with 301 redirects
• ✅ Hierarchical TOC, citation copy, enhanced accessibility

Phase 10-11: Professional Polish & User Magnet - ✅ COMPLETE

• ✅ SEO hardening (canonical URLs, alternates, expanded sitemap, OG images)
• ✅ Cross-reference panel, version timeline
• ✅ 6-format export (TXT/PDF/LaTeX/DOCX/EPUB/JSON) with tier-based rate limits
• ✅ Word-level compare diff
• ✅ Cmd+K global search overlay, citation + BibTeX export
• ✅ Homepage refresh (FeaturedLaws, QuickLinks, trilingual headings)

Phase 12: Full Legal Framework Data Gathering - ✅ COMPLETE

• ✅ State scrapers (BC, Durango, QR) — 1,102 additional laws
• ✅ Municipal scrapers (GDL, MTY, León, Zapopan) — 2,195 files
• ✅ NOM catalog (80 DOF NOMs), OCR pipeline fallback

Phase 13: API Platform — Ecosystem & External Customers - ✅ COMPLETE

• ✅ API key infrastructure (generation, hashing, auth backend, combined auth chain)
• ✅ Tiered rate limiting + usage logging middleware
• ✅ Bulk data feed (/bulk/articles/) + changelog (/changelog/) endpoints
• ✅ Webhook subscriptions with HMAC-SHA256 signing + auto-disable
• ✅ Domain filtering (?domain=finance) on all endpoints
• ✅ TypeScript SDK (@tezca/api-client) with auto-pagination
• ✅ API key admin CRUD + CORS for API key consumers
• ✅ OpenAPI security schemes (ApiKey + Bearer)

Phase 14: Production & Growth - 🔄 IN PROGRESS

• ✅ Production go-live at tezca.mx (web + api.tezca.mx live behind Cloudflare)
• 📋 Embeddings / vector search integration
• 📋 External customer onboarding (legaltech, compliance SaaS)

See: ROADMAP.md for detailed timeline and Strategic Overview for comprehensive vision

Contributing

See CONTRIBUTING.md for guidelines.

License

AGPL-3.0 — see pyproject.toml for details.

Contact

Issues: https://github.com/madfam-org/tezca/issues