list_sources
Retrieve available data sources from the Unstructured API, optionally filtered by connector type, to identify integration options for document processing workflows.
Instructions
List available sources from the Unstructured API.
Args:
source_type: Optional source connector type to filter by
Returns:
String containing the list of sourcesInput Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| source_type | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- uns_mcp/server.py:113-152 (handler)The main handler function for the 'list_sources' MCP tool, decorated with @mcp.tool(). It retrieves the list of source connectors from the Unstructured API client, optionally filters by source_type, sorts them, and returns a formatted string listing the sources with their names and IDs.
@mcp.tool() async def list_sources( ctx: Context, source_type: Optional[SourceConnectorType | str] = None, ) -> str: """ List available sources from the Unstructured API. Args: source_type: Optional source connector type to filter by Returns: String containing the list of sources """ client = ctx.request_context.lifespan_context.client request = ListSourcesRequest() if source_type: try: source_type = ( SourceConnectorType(source_type) if isinstance(source_type, str) else source_type ) request.source_type = source_type except KeyError: return f"Invalid source type: {source_type}" response = await client.sources.list_sources_async(request=request) # Sort sources by name sorted_sources = sorted(response.response_list_sources, key=lambda source: source.name.lower()) if not sorted_sources: return "No sources found" # Format response result = ["Available sources:"] for source in sorted_sources: result.append(f"- {source.name} (ID: {source.id})") return "\n".join(result)