### Install and Run MCP Server (Bash)

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Instructions for cloning the repository, installing dependencies using UV, and running the MCP server locally.

```bash
# Clone the repository
git clone https://github.com/appunite/dane-gov-pl-mcp.git
cd dane-gov-pl-mcp

# Install dependencies
uv sync

# Run the MCP server
uv run python -m src.app --transport stdio
```

--------------------------------

### Discovery API Usage (Python)

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Python examples demonstrating how to use the Discovery layer of the MCP server to search for datasets, institutions, and retrieve resource details.

```python
# Search datasets by keywords
search_datasets(search_filters={"query_all": "environment"})

# Find institutions by location
search_institutions(search_filters={"city_terms": "Warszawa"})

# Get dataset details
get_resources_details(dataset_ids=[123, 456])
```

--------------------------------

### Document Parsing API Usage (Python)

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Python example for using the Parsing layer of the MCP server to convert various file formats (CSV, JSON, XLSX, PDF) into Markdown documents.

```python
# Parse files to Markdown
get_file_content(resource_ids=[123, 456])
```

--------------------------------

### MCP Client Configuration (JSON)

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Configuration snippets for integrating the Dane.gov.pl MCP Server with MCP clients like Claude Desktop or Cursor. Includes options for UV package manager.

```json
{
  "mcpServers": {
    "dane-gov-pl-mcp": {
      "command": "uvx",
      "args": ["dane-gov-pl-mcp"]
    }
  }
}
```

```json
{
  "mcpServers": {
    "dane-gov-pl-mcp": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/dane-gov-pl-mcp", "dane-gov-pl-mcp"]
    }
  }
}
```

--------------------------------

### Discovery API

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Endpoints for searching and filtering government institutions, datasets, resources, and showcases.

```APIDOC
## Discovery API Endpoints

### Search Datasets

**Description**: Search for datasets using keywords and various filters.

**Method**: GET (assumed, based on usage example)

**Endpoint**: `/datasets` (assumed)

**Query Parameters**:
- **query_all** (string) - Optional - Keywords to search across all dataset fields.
- **title_terms** (string) - Optional - Filter by dataset title.
- **description_terms** (string) - Optional - Filter by dataset description.

### Search Institutions

**Description**: Find and filter government institutions by name, city, and description.

**Method**: GET (assumed, based on usage example)

**Endpoint**: `/institutions` (assumed)

**Query Parameters**:
- **city_terms** (string) - Optional - Filter by city name.
- **name_terms** (string) - Optional - Filter by institution name.
- **description_terms** (string) - Optional - Filter by institution description.

### Get Dataset Resource Details

**Description**: Retrieve detailed information about specific data resources within datasets.

**Method**: GET (assumed, based on usage example)

**Endpoint**: `/datasets/{dataset_id}/resources` (assumed)

**Query Parameters**:
- **dataset_ids** (list of integers) - Required - A list of dataset IDs to retrieve resource details for.

### Search Showcases

**Description**: Find real-world visualizations and applications that use the datasets.

**Method**: GET (assumed, based on usage example)

**Endpoint**: `/showcases` (assumed)

**Query Parameters**:
- **query_all** (string) - Optional - Keywords to search across showcase fields.

### Request Example (Python - Search Datasets)
```python
# Assuming a client library is available
from mcp_client import search_datasets

results = search_datasets(search_filters={"query_all": "environment"})
print(results)
```

### Request Example (Python - Search Institutions)
```python
# Assuming a client library is available
from mcp_client import search_institutions

results = search_institutions(search_filters={"city_terms": "Warszawa"})
print(results)
```

### Request Example (Python - Get Dataset Resource Details)
```python
# Assuming a client library is available
from mcp_client import get_resources_details

results = get_resources_details(dataset_ids=[123, 456])
print(results)
```
```

--------------------------------

### Parsing and Processing API

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Endpoints for parsing various file formats into Markdown documents and performing LLM-powered operations on tabular data.

```APIDOC
## Parsing and Processing API Endpoints

### Get File Content (Parsing)

**Description**: Convert resources (files) into LLM-ready Markdown documents.

**Method**: GET (assumed, based on usage example)

**Endpoint**: `/resources/{resource_id}/content` (assumed)

**Query Parameters**:
- **resource_ids** (list of integers) - Required - A list of resource IDs to parse.

**Response Example (Markdown Document)**:
```markdown
# Dataset Title

## Description

This is a description of the dataset.

## Data Table

| Header 1 | Header 2 |
|---|---|
| Row 1 Col 1 | Row 1 Col 2 |
| Row 2 Col 1 | Row 2 Col 2 |
```

### Tabular Data Processing

**Description**: Enable grouping, aggregating, filtering, and sorting operations for tabular data resources. This functionality is typically accessed through a client that sends commands to the MCP server to perform these operations on data loaded into Polars DataFrames.

**Method**: POST (assumed for operations)

**Endpoint**: `/process` (assumed)

**Request Body Example (Conceptual)**:
```json
{
  "operation": "aggregate",
  "data_source": {
    "type": "resource",
    "id": 123
  },
  "group_by": "category",
  "aggregations": {
    "value": "sum"
  }
}
```

**Response Example (Processed Data - Conceptual)**:
```json
{
  "result": [
    {"category": "A", "value": 150},
    {"category": "B", "value": 200}
  ]
}
```

### Request Example (Python - Get File Content)
```python
# Assuming a client library is available
from mcp_client import get_file_content

markdown_docs = get_file_content(resource_ids=[123, 456])
for doc in markdown_docs:
    print(doc)
```
```

--------------------------------

### Retrieve Tabular Resource Metadata

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Fetches metadata for specific tabular resources using their unique identifiers. This function is essential for understanding the structure and availability of datasets before querying them.

```python
get_tabular_resource_metadata(resource_ids=[123])
```

--------------------------------

### Query Tabular Data

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Executes a query against a specific tabular resource using search filters. It allows users to retrieve filtered subsets of data based on column-specific criteria.

```python
get_tabular_data(resource_id=123, search_filters={"q": "col1:Warszawa"})
```

--------------------------------

### Perform Advanced DataFrame Operations

Source: https://github.com/appunite/dane-gov-pl-mcp/blob/main/README.md

Transforms and aggregates data from a resource into a DataFrame format. It supports complex operations such as grouping, multi-column aggregation, and multi-level sorting.

```python
resource_to_dataframe(resource_id=123, dataframe_operations={
	"primary_group": "col1",
	"aggregations": ["sum", "mean"],
	"aggregation_columns": ["col2", "col2"],
	"sort_columns": ["col2_sum", "col1"],
	"sort_descending": [True, False]
})
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.