What are good ways to handle software that has an ambiguous license version like 'AGPL'?
License Naming Background
FSF is considered the steward of AGPL version 3.0 (https://www.fsf.org/news/agplv3-pr , https://opensource.org/blog/steward/free-software-foundation) and provides interpretations around GNU licenses (https://www.fsf.org/blogs/executive-director/a-short-update-on-gnu-general-public-license-gpl-compatibility-questions).
GNU considers AGPLv1.0 non-open source while AGPLv3.0 is open source: https://www.gnu.org/licenses/license-list.en.html. GNU's page lists short forms for license names with AGPL and AGPLv3.0 both referring to version 3.0.
SPDX has four recognized short forms for AGPL that are all non-ambigous (https://spdx.org/licenses/): AGPL-1.0-only, AGPL-1.0-or-later, AGPL-3.0-only, AGPL-3.0-or-later.
OSI recognizes AGPL version 3.0 as open source but only recognizes the SPDX short form of AGPL-3.0-only apparently not extending its approval to AGPL-3.0-or-later. I guess it would be presumptuous to hold that another organization will always meet certain criteria when publishing a new version.
If I understand correctly from private communications, FSF and GNU do not subscribe to SPDX, in part because full license text is considered authoritative, not a short identifier.
Concrete Scenario
In an open source community around an AGPLv3.0 source code with an AFLv3.0 contribution policy (https://civicrm.org/about/license) there are a few thousand extensions. The extension system includes a parameter in an xml file for the license of the extension, but there is no validation of the value specified by a developer creating the xml file. Several dozen of these extensions specify AGPL as the license, and some of these do not include a LICENSE file in their repo. Is it reasonable to assume this software is legally AGPL version 3.0? Is it reasonable to consider it AGPL version 3.0 only, or AGPL version 3.0 or later?