Kaltura MCP Server

Instructions

Implementation Reference

• src/kaltura_mcp/tools/search.py:39-82 (handler)

  The core handler function that executes the list_categories tool: filters categories by optional search text, paginates results, extracts key metadata, and returns formatted JSON response.

  async def list_categories(
      manager: KalturaClientManager,
      search_text: Optional[str] = None,
      limit: int = 20,
  ) -> str:
      """List available categories."""
      client = manager.get_client()
  
      # Create filter
      filter = KalturaCategoryFilter()
      if search_text:
          filter.freeText = search_text
  
      # Create pager
      pager = KalturaFilterPager()
      pager.pageSize = limit
  
      # List categories
      result = client.category.list(filter, pager)
  
      categories = []
      for category in result.objects:
          categories.append(
              {
                  "id": category.id,
                  "name": category.name,
                  "description": category.description,
                  "tags": category.tags,
                  "fullName": category.fullName,
                  "depth": category.depth,
                  "entriesCount": category.entriesCount,
                  "createdAt": datetime.fromtimestamp(category.createdAt).isoformat()
                  if category.createdAt
                  else None,
              }
          )
  
      return json.dumps(
          {
              "totalCount": result.totalCount,
              "categories": categories,
          },
          indent=2,
      )

• src/kaltura_mcp/server.py:87-103 (schema)

  MCP tool schema definition including input validation for search_text and limit parameters, with detailed usage description.

  types.Tool(
      name="list_categories",
      description="Browse content organization hierarchy. USE WHEN: Exploring content structure, finding category IDs for filtering, understanding content taxonomy. Categories organize videos into folders/topics. RETURNS: Tree structure with category names, IDs, parent-child relationships. EXAMPLE: Find all videos in 'Training' category by first getting category ID.",
      inputSchema={
          "type": "object",
          "properties": {
              "search_text": {
                  "type": "string",
                  "description": "Optional text to filter categories (e.g., 'training', 'marketing')",
              },
              "limit": {
                  "type": "integer",
                  "description": "Maximum number of categories to return (default: 20)",
              },
          },
      },
  ),

• src/kaltura_mcp/server.py:499-500 (registration)

  Tool dispatch registration in the MCP server's call_tool handler, invoking the list_categories function with Kaltura manager and arguments.

  elif name == "list_categories":
      result = await list_categories(kaltura_manager, **arguments)

• src/kaltura_mcp/tools/__init__.py:28-32 (registration)

  Package-level import registration exposing list_categories from the search module to the tools namespace.

      esearch_entries,
      list_categories,
      search_entries,
      search_entries_intelligent,
  )