INEI MCP Server
[](https://lobehub.com/mcp/josetra44-inei-mcp)
[](LICENSE)
[](https://python.org)
[]()
# INEI MCP Server
Connects any MCP-compatible AI agent to **INEI Estadist** — the official statistics platform of Peru's [Instituto Nacional de Estadística e Informática (INEI)](https://www.inei.gob.pe/).
Access Census 2017 data, population indicators, geographic profiles, and social statistics for all 25 departments, 196 provinces, and 1,874 districts of Peru — **no API key required**.
**Works with:** Claude Desktop · Claude Code · Cursor · VS Code Copilot · Windsurf · any MCP client
---
## Quickstart
```bash
# From source (development)
git clone https://github.com/JOSETRA44/inei-mcp.git
cd inei-mcp
uv sync
uv run inei-mcp
```
---
## Available Tools (7)
| Tool | Description |
|------|-------------|
| `inei_get_departments` | List all 25 Peruvian departments with ubigeo codes and IDs |
| `inei_get_provinces` | List provinces of a department by ccdd code |
| `inei_get_districts` | List districts of a province, or search districts by name |
| `inei_search_indicators` | Search statistical indicators by keyword |
| `inei_get_indicator_data` | Get indicator values across geographies |
| `inei_browse_topics` | Browse the full thematic indicator tree |
| `inei_get_geography_profile` | Census 2017 population profile for a region |
| `inei_get_census_dashboard` | National demographic or social dashboard |
| `inei_get_data_sources` | List official data sources |
---
## Available Prompts (3)
| Prompt | Description |
|--------|-------------|
| `demographic_analysis` | Structured demographic analysis for a department |
| `regional_comparison` | Compare an indicator across Peru's regions |
| `census_deep_dive` | Comprehensive Census 2017 analysis |
---
## Available Resources (3)
| URI | Contents |
|-----|----------|
| `inei://api-guide` | Complete API reference |
| `inei://department-codes` | All 25 department codes and IDs |
| `inei://thematic-categories` | Indicator categories with example IDs |
---
## Configuration (Claude Code / Claude Desktop)
### From source (development)
```json
{
"mcpServers": {
"inei": {
"command": "uv",
"args": ["--directory", "C:\\Users\\USER\\source\\MCPs\\INEI-mcp", "run", "inei-mcp"]
}
}
}
```
---
## Data Source
All data from **INEI Estadist** — Peru's official national statistics platform.
Primary data: **Censos Nacionales 2017** — XII de Población y VII de Vivienda.
---
## License
[MIT](LICENSE) — © 2026 JOSETRA44
TDQS
Scored across 9 tools
The tool set has a major disambiguation problem due to inei_get_districts being misnamed: its description says it returns all 25 departments, not districts, making it redundant with inei_get_departments. Other tools like inei_browse_topics, inei_search_indicators, and inei_get_indicator_data have distinct roles, but the duplication causes confusion.
All tools use the 'inei_' prefix and snake_case, with a consistent verb_noun pattern (get_*, browse_*, search_*). However, inei_get_districts is misleading because it does not return districts but departments, breaking semantic consistency. inei_get_census_dashboard uses 'dashboard' instead of a data-object noun, which is a minor deviation.
With 9 tools, the set is well-scoped for a statistical API covering browsing, searching, geography listing, geography profiling, and data retrieval. Each tool has a clear purpose, and the count feels neither too sparse nor overwhelming.
The set covers the main workflows: explore topics, search indicators, retrieve data by geography, and get national summaries. It can handle department and province queries. However, the duplicate get_districts tool is unnecessary, and the API's limitation to department-level geography (no district data) is a domain constraint. Minor gaps exist (e.g., no hierarchical drill-down beyond two levels), but overall coverage is good for a read-only census API.