Implement comprehensive NDP MCP server for dataset search and discovery

This implementation provides a robust Model Context Protocol (MCP) server for the National Data Platform API, enabling seamless dataset search and discovery across multiple CKAN instances.

Key features:
- Multi-server support (local, global, pre-production CKAN instances)
- Organization discovery with filtering capabilities
- Comprehensive dataset search (simple term-based and advanced field-specific)
- Dataset detail retrieval with complete metadata
- Robust error handling with retry logic for network failures
- Automatic result limiting to prevent context overflow (default: 20 results)
- String/integer parameter handling for MCP client compatibility

Tools implemented:
- list_organizations: Discover available data organizations
- search_datasets: Search datasets with flexible filtering options
- get_dataset_details: Retrieve complete dataset metadata by ID or name

Technical improvements:
- FastMCP framework integration with proper type annotations
- Pydantic models for data validation (Dataset model)
- Comprehensive test suite with 30 tests covering edge cases
- UV package manager integration for modern Python workflow
- Extensive documentation following IOWarp MCP standards

The server addresses common issues like token overflow by implementing smart limiting and provides clear workflow guidance for effective dataset discovery.

Author: JaimeCernuda

mcps/NDP/README.md

@@ -0,0 +1,220 @@
+# API Search Documentation
+
+This document provides comprehensive information about the search functionality in the API, specifically covering the `/organizations` and `/search` endpoints.
+
+## Organizations Endpoint
+
+### GET /organization - List Organizations
+
+**Purpose**: Retrieve a list of all organizations with optional filtering.
+
+**Method**: `GET`
+
+**Parameters**:
+- `name` (query, optional): Filter organizations by name
+- `server` (query, optional): Specify server to list organizations from
+  - Options: `local`, `global`, `pre_ckan`
+  - Default: `global`
+
+**Response**:
+- **200 OK**: Returns an array of organization names
+- **400 Bad Request**: Error message explaining the bad request
+- **422 Validation Error**: Validation error details
+
+**Example Request**:
+```
+GET /organization?name=example&server=global
+```
+
+**Example Response**:
+```json
+["org1", "org2", "org3"]
+```
+
+### POST /organization - Create Organization
+
+**Purpose**: Create a new organization.
+
+**Method**: `POST`
+
+**Parameters**:
+- `server` (query, optional): Specify server (`local` or `pre_ckan`, defaults to `local`)
+
+**Request Body**:
+```json
+{
+  "name": "example_org_name",           // Required: Unique organization name
+  "title": "Example Organization Title", // Required: Organization title
+  "description": "This is an example organization." // Optional: Description
+}
+```
+
+**Response**:
+- **201 Created**: Organization created successfully
+- **400 Bad Request**: Organization name already exists
+- **422 Validation Error**: Validation error details
+
+**Example Response**:
+```json
+{
+  "id": "305284e6-6338-4e13-b39b-e6efe9f1c45a",
+  "message": "Organization created successfully"
+}
+```
+
+## Search Endpoint
+
+### GET /search - Search Datasets by Terms
+
+**Purpose**: Search CKAN datasets using a list of search terms.
+
+**Method**: `GET`
+
+**Parameters**:
+- `terms` (query, required): Array of search terms
+- `keys` (query, optional): Array of keys corresponding to each term (use `null` for global search)
+- `server` (query, optional): Server to search on
+  - Options: `local`, `global`
+  - Default: `global`
+  - Note: If 'local' CKAN is disabled, it cannot be used
+
+**Response**:
+- **200 OK**: Returns array of matching datasets
+- **400 Bad Request**: Error message explaining the bad request
+- **422 Unprocessable Entity**: Validation error details
+
+**Example Request**:
+```
+GET /search?terms=climate&terms=data&server=global
+```
+
+### POST /search - Advanced Dataset Search
+
+**Purpose**: Search datasets using various parameters with more granular control.
+
+**Method**: `POST`
+
+**Request Body Schema**:
+All fields are optional and can be used in combination:
+
+#### Common Registration-Matching Parameters:
+- `dataset_name`: The name of the dataset
+- `dataset_title`: The title of the dataset
+- `owner_org`: The name of the organization
+- `resource_url`: The URL of the dataset resource
+- `resource_name`: The name of the dataset resource
+- `dataset_description`: The description of the dataset
+- `resource_description`: The description of the resource
+- `resource_format`: The format of the dataset resource
+
+#### User-Defined Search Parameters:
+- `search_term`: Comma-separated list of terms to search across all fields
+- `filter_list`: Array of field filters in the form `key:value`
+- `timestamp`: Filter on the timestamp field of results
+- `server`: Server selection (`local`, `global`, or `pre_ckan`, defaults to `global`)
+
+**Example Request Body**:
+```json
+{
+  "dataset_name": "climate_data",
+  "resource_format": "CSV",
+  "search_term": "temperature,weather",
+  "filter_list": ["type:sensor", "location:europe"],
+  "server": "global"
+}
+```
+
+**Response**:
+- **200 OK**: Returns array of matching datasets
+- **400 Bad Request**: Error occurred during search
+- **422 Validation Error**: Request validation failed
+
+## Response Schema
+
+Both search endpoints return an array of `DataSourceResponse` objects:
+
+```json
+[
+  {
+    "id": "12345678-abcd-efgh-ijkl-1234567890ab",
+    "name": "example_dataset_name",
+    "title": "Example Dataset Title",
+    "owner_org": "example_org_name",
+    "notes": "This is an example dataset.",
+    "resources": [
+      {
+        "id": "abcd1234-efgh5678-ijkl9012",
+        "url": "http://example.com/resource",
+        "name": "Example Resource Name",
+        "description": "This is an example resource.",
+        "format": "CSV"
+      }
+    ],
+    "extras": {
+      "key1": "value1",
+      "key2": "value2",
+      "mapping": {
+        "field1": "qeadw2",
+        "field2": "gw4aw34",
+        "time": "gw4aw34"
+      },
+      "processing": {
+        "data_key": "",
+        "info_key": "key_with_info"
+      }
+    }
+  }
+]
+```
+
+### DataSource Response Fields:
+- `id` (string, required): Unique dataset identifier
+- `name` (string, required): Unique dataset name
+- `title` (string, required): Dataset title
+- `owner_org` (string, optional): Organization ID that owns the dataset
+- `notes` (string, optional): Dataset description
+- `resources` (array, required): List of associated resources
+- `extras` (object, optional): Additional metadata
+
+### Resource Object Fields:
+- `id` (string, required): Unique resource identifier
+- `url` (string, optional): Resource URL
+- `name` (string, required): Resource name
+- `description` (string, optional): Resource description
+- `format` (string, optional): Resource format (e.g., CSV, JSON, etc.)
+
+## Search Examples
+
+### 1. Simple term search:
+```bash
+curl -X GET "http://155.101.6.191:8003/search?terms=climate&server=global" \
+  -H "Authorization: Bearer <token>"
+```
+
+### 2. Advanced search with multiple criteria:
+```bash
+curl -X POST "http://155.101.6.191:8003/search" \
+  -H "Content-Type: application/json" \
+  -H "Authorization: Bearer <token>" \
+  -d '{
+    "dataset_name": "sensor_data",
+    "resource_format": "CSV",
+    "search_term": "temperature,humidity",
+    "server": "global"
+  }'
+```
+
+### 3. List organizations with filtering:
+```bash
+curl -X GET "http://155.101.6.191:8003/organization?name=research&server=global" \
+  -H "Authorization: Bearer <token>"
+```
+
+## Server Options
+
+The API supports multiple server configurations:
+- **local**: Search in local CKAN instance (may be disabled)
+- **global**: Search in global CKAN instance (default for most endpoints)
+- **pre_ckan**: Search in pre-production CKAN instance (available for some endpoints)
+
+Choose the appropriate server based on your data access requirements and instance availability.