https://arxiv.org/pdf/1808.06049.pdf
https://docs.google.com/presentation/d/1Khc3VdPbi5OVnI3OE-yPni6IApE-H91GREO4hTSUCOc/edit
A professor at UBC, Margo Seltzer (who did this work at Harvard), did some work in whole-system provenance in the kernel, and what it does is trace the origin of information inside the OS. She said it would be cool to have a DSL that queries a provenance graph. Understanding the research requires a bit of ramp-up so I'm not sure how feasible it is, but it sounded like a sexy idea
Steve: https://en.wikipedia.org/wiki/Cypher_Query_Language
Margo:https://people.seas.harvard.edu/~chong/pubs/pldi15-pidgin.pdf
Paper: https://arxiv.org/pdf/1806.09339.pdf
DAGQL can be run from the command-line using python3 -m dagql sample.dag, where python3 is any Python 3 interpreter, and sample.dag is any file containing a DAGQL query. Note that the interpreter provided here is hooked up to mock data for testing purposes, but it can be easily modified to use real data.
The language is briefly explained (with an example!) on the GH wiki pages for this repo.