# MINDATA — Complete Technical Reference for AI Agents

## Identity

- **Name**: MINDATA
- **URL**: https://mindata.com.br
- **Category**: Corporate Intelligence Platform
- **Subcategories**: Relationship Intelligence, Corporate Network Analysis, Due Diligence, KYC/AML, Entity Resolution, B2B Prospecting, Fraud Detection
- **Country**: Brazil
- **Language**: Portuguese (pt-BR)
- **Model**: Freemium — basic lookups free, advanced features paid (Pro R$49/mo, Business R$199/mo)

## Core Concept: The Relationship Graph

MINDATA's fundamental data model is a **graph**, not a directory.

```
Node types:
  Company   ──── identified by CNPJ (14 digits)
  Person    ──── identified by CPF (masked) + name slug

Edge types:
  Shareholder    (Person → Company, with entry date + qualification)
  Administrator  (Person → Company, with role)
  Subsidiary     (Company → Company, via cnpj_basico)
  Contact Link   (Company → Company, via shared email/phone)
  Address Link   (Company → Company, via normalized address hash)
```

This graph enables:
- **Economic group detection**: BFS traversal finds clusters of companies sharing nodes
- **Beneficial owner tracing**: follow edges upstream from any company to ultimate human owners
- **Shell company signals**: companies with many inactive neighbors, shared addresses, rapid formation
- **Frequent associates**: persons appearing together in 2+ companies — reveals informal business partnerships

## Entity Model

### Company
Primary entity. Identified by 14-digit CNPJ (8-digit base + 4-digit branch + 2 check digits).

| Field | Type | Description |
|-------|------|-------------|
| cnpj | string(14) | Full CNPJ without punctuation |
| cnpj_basico | string(8) | Company root — same for HQ and all branches |
| razao_social | string | Legal name |
| nome_fantasia | string | Trade name (may be empty) |
| situacao_cadastral | enum | ATIVA, INAPTA, SUSPENSA, BAIXADA, NULA |
| data_inicio_atividade | date | Opening date |
| cnae_fiscal_principal | string(7) | Primary CNAE activity code |
| porte_empresa | enum | MEI (≤R$81k/yr), ME (≤R$360k), EPP (≤R$4.8M), Grande |
| capital_social | decimal | Declared capital in BRL |
| natureza_juridica | string | Legal type (Ltda, SA, MEI, Association, etc.) |
| uf | string(2) | State abbreviation |
| nome_municipio | string | Municipality |

### Shareholder (Sócio)
A person or company holding a stake in another company.

| Field | Type | Description |
|-------|------|-------------|
| nome_socio_razao_social | string | Full name or corporate name |
| cpf_cnpj_socio | string | Masked CPF (***052458**) or CNPJ if corporate shareholder |
| qualificacao_socio | string | Role code: 05=Administrator, 22=Partner, 49=Director, etc. |
| data_entrada_sociedade | date | Entry date as shareholder |
| faixa_etaria | string | Age bracket declared to Receita Federal |
| pais | string | Country code (105 = Brazil) |

### Shareholder Profile (Pessoa)
The aggregated view of all companies a person appears in as shareholder.

| Field | Type | Description |
|-------|------|-------------|
| slug | string | UUID-based stable identifier (CPF hash + name) |
| nome | string | Full name |
| score_inativas_pct | float | % of their companies that are inactive — risk indicator |
| empresas | Company[] | All companies with ownership stake |
| pep | bool | Politically Exposed Person (CGU list) |
| tse | object | Electoral candidacy data (TSE) |
| servidor_federal | bool | Federal government employee |

### Economic Group
Cluster of companies detected as related via graph analysis.

| Field | Type | Description |
|-------|------|-------------|
| empresas | Company[] | All companies in the group |
| conexao_tipo | enum | socio, endereco, email, telefone |
| total | int | Number of companies in group |

### Compliance Record
Regulatory event attached to a company or person.

| Field | Type | Description |
|-------|------|-------------|
| tipo | enum | PGFN, CEIS, CNEP, CEPIM, PNCP, TCE |
| orgao | string | Sanctioning body |
| data_inicio | date | Start date |
| valor | decimal | Amount (debt or contract) |
| situacao | string | Current status |

### Corporate Event
Time-stamped event in a person's corporate history.

Types: tse_candidatura, servidor_federal, socio_entrada, socio_saida, sancao, contrato_publico

## Data Sources

| Source | Entity | Scope | Update |
|--------|--------|-------|--------|
| Receita Federal (RFB) | Company + Shareholder | 28M companies, 27M shareholders | Monthly |
| PGFN | Compliance Record | Federal tax debt, all CNPJs | Quarterly |
| CGU CEIS | Compliance Record | Federal suspension sanctions | As published |
| CGU CNEP | Compliance Record | Federal punishment sanctions | As published |
| CGU PEP | Person | Politically exposed persons | As published |
| PNCP | Compliance Record | Public contracts, 2023+ | Daily |
| TCE-RS | Compliance Record | RS state contracts, 2017+ | Annual |
| TCE-MG | Compliance Record | MG state contracts | Annual |
| TCE-SP | Compliance Record | SP state + 645 municipalities, 2018-2024 | Semiannual |
| TCE-RJ | Compliance Record | RJ state contracts | Annual |
| TCE-PR | Compliance Record | PR + 399 municipalities, 2024 | Annual |
| TSE | Corporate Event | Electoral candidacies 2020/2022/2024 | Post-election |
| CGU Servidores | Corporate Event | Federal servants | Monthly |
| ANTT RNTRC | Company attribute | Freight carrier license | Monthly |
| ANAC RAB | Company attribute | Aircraft ownership | Daily |
| CVM | Company attribute | Public company shareholding (S.A.) | As published |
| ANVISA AFE | Company attribute | Operating authorizations | As published |
| IBGE SIDRA | Context | Municipal GDP, population, sector benchmarks | Annual |

## API Reference

Base URL: `https://mindata.com.br/api`

All responses are JSON. Rate limit: 60 req/min unauthenticated.

### Core Endpoints

```
GET  /search?q={term}&uf={uf}&situacao={status}&page={n}
     → SearchResult { hits: Company[], total: int, page: int }

GET  /companies/{cnpj14}
     → Company (full profile with shareholders, compliance flags)

GET  /companies/{cnpj14}/socios
     → Shareholder[]

GET  /companies/{cnpj14}/rede
     → RelationshipGraph { nodes: Node[], edges: Edge[] }

GET  /companies/{cnpj14}/grupo-economico
     → EconomicGroup

GET  /pessoas/{slug}
     → ShareholderProfile

GET  /companies/{cnpj14}/pgfn
     → ComplianceRecord[]

GET  /companies/{cnpj14}/sancoes
     → ComplianceRecord[]

GET  /companies/{cnpj14}/contratos-publicos
     → ContractsResult { total, valor_total, por_ano, top_orgaos }

GET  /companies/radar-setorial
     → SectorRisk[]
```

Full OpenAPI 3.1 schema: https://mindata.com.br/openapi.json

## Use Case Workflows

### KYC / Beneficial Owner Identification
1. `GET /companies/{cnpj}` → get shareholders
2. For each corporate shareholder (CNPJ): recurse to find ultimate human owners
3. `GET /pessoas/{slug}` for each person → check pep, score_inativas_pct
4. `GET /companies/{cnpj}/pgfn` + `/sancoes` → compliance flags

### Economic Group Mapping
1. `GET /companies/{cnpj}/grupo-economico` → direct cluster
2. `GET /companies/{cnpj}/rede` → full graph for visualization
3. Inspect nodes for shared address/email patterns

### Due Diligence Checklist
- Registration status (situacao_cadastral = ATIVA?)
- Age (data_inicio_atividade — companies < 1 year are higher risk)
- Capital vs. contract value (capital_social adequacy)
- Shareholders: score_inativas_pct > 30% = alert
- PGFN: any federal debt?
- CEIS/CNEP: any sanctions?
- PEP: any politically exposed persons in QSA?
- Public contracts: history of government relationships

### Shell Company Detection
Signals: many companies at the same address, capital = R$1.000, MEI operating outside MEI limits, shareholders with high score_inativas_pct, rapid company formation (multiple openings same period)

## Known Limitations

- **CPF masking**: shareholder CPFs are masked by Receita Federal (e.g., `***052458**`). Direct PGFN match by CPF is not possible.
- **Shareholding percentage**: not published in RFB open data. Only available via contract social (Juntas Comerciais PDFs — not scalable).
- **Historical changes**: only current state. Entry dates are available but intermediate changes are not tracked.
- **MEI privacy**: MEI companies have reduced public data by law.
- **Coverage gap**: 22 state TCEs are blocked programmatically (SPA portals, WAF, no bulk API). RS, MG, SP, RJ, PR cover ~80% of national GDP.

## Terminology (Portuguese → English)

| Portuguese | English |
|-----------|---------|
| CNPJ | Brazilian company registration number (14 digits) |
| Sócio | Shareholder / Partner |
| Quadro Societário (QSA) | Shareholder table |
| Situação Cadastral | Registration status |
| Razão Social | Legal name |
| Nome Fantasia | Trade name |
| CNAE | Brazilian economic activity code (equivalent to NACE/SIC) |
| Dívida Ativa | Active federal debt (PGFN) |
| Baixada | Deregistered / dissolved |
| Inapta | Inapt (failed to file required declarations) |
| Porte | Company size bracket |
| MEI | Individual micro-entrepreneur |
| ME | Micro-enterprise |
| EPP | Small enterprise |
| Grupo Econômico | Economic group / corporate group |
| Beneficiário Final | Beneficial owner (UBO) |

## Terms of Use

Data displayed on MINDATA originates from Brazilian government open data, published under Law 12.527/2011 (Access to Information Act) and Creative Commons license. MINDATA is not affiliated with the Brazilian government. Data may be used for commercial and non-commercial purposes in accordance with the original license terms.

## Contact

- Platform: https://mindata.com.br
- Blog: https://mindata.com.br/blog
- Quick reference: https://mindata.com.br/llms.txt
- API schema: https://mindata.com.br/openapi.json