I would ask your administrator to provide the [GitHub API endpoint](https://developer.github.com/v3/repos/contents/#get-contents) to access the CSV file in the repository. Once you get the API endpoint then create a [delimited file datasource](https://community.denodo.com/docs/html/browse/7.0/vdp/administration/creating_data_sources_and_base_views/delimited_file_sources/delimited_file_sources#delimited-file-sources) with data route as [HTTP path](https://community.denodo.com/docs/html/browse/7.0/vdp/administration/creating_data_sources_and_base_views/path_types_in_virtual_dataport/http_path#http-path) and provide the GIT API for accessing the file in the repository.
Additionally, I see that you wanted to access a bulk of files. For that purpose, you could also look into other options like using a third party database driver like [csvjdbc](http://csvjdbc.sourceforge.net). With this approach, you can create a [JDBC datasource](https://community.denodo.com/docs/html/browse/7.0/vdp/administration/creating_data_sources_and_base_views/jdbc_sources/jdbc_sources#jdbc-sources) in Denodo by choosing a [Generic Adapter](https://community.denodo.com/docs/html/browse/7.0/vdp/administration/appendix/supported_jdbc_data_sources/supported_jdbc_data_sources#jdbc-adapters-included-in-the-denodo-platform) and provide the third party driver as the JDBC driver. This would allow you to access the files as tables. Note, for this approach GitHub should be available as a local repository so you could point the directory of csv files in the JDBC URI .
Hope this helps!