Headscale QA test using Claude AI|.claude/agents/headscale-integration-tester.md

This is interesting, as we're looking at testing methodology defined with a good level of detail, with prohibitive rules, such that an AI can execute it, which arguably makes it better documentation than most teams have for their human developers.

I also think it is interesting to use an LLM because it will likely be a bit more human than a normal integration automated test.