Project Graphs
Build method, file, and module dependency views from real Julia projects with one local analysis workflow.
These small examples are rendered from real DependencyAtlas analysis output and kept static for the docs site.
struct Dog end; struct Cat end;
meet(::Dog, ::Dog) = play_together()
meet(::Dog, ::Cat) = chase_cat()
meet(::Cat, ::Dog) = hiss_at_dog()
meet(::Cat, ::Cat) = ignore_each_other()
play_together() = "The dogs play together"
chase_cat() = "The dog chases the cat"
hiss_at_dog() = "The cat hisses at the dog"
ignore_each_other() = "The cats ignore each other"
demo() = meet(Dog(), Cat())
module MiniDependencyAtlas
include("analysis.jl")
include("graph.jl")
include("query.jl")
include("runtime.jl")
include("export.jl")
run_demo(path::String) =
serve_snapshot(export_graph(query_graph(build_graph(path))))
end
We are collecting good-looking theme presets; if you have one you would like to share, export it as JSON and send it through an issue or pull request. It may become a built-in DependencyAtlas theme.






Build method, file, and module dependency views from real Julia projects with one local analysis workflow.
Explore callers, callees, and bounded traversal results without being trapped by the current visible graph.
Open a selected graph node directly in your editor at the exact source line behind the node.
Analysis runs locally, so source code and graph data stay on your machine by default.
The default product path is fast and source-owned. JET stays available as an opt-in view when a selected method needs compiler-time evidence.
Builds the main snapshot and owns the canonical source facts: files, modules, methods, includes, imports, overload sites, and lightweight static call evidence.
Runs from a selected method node when deeper compiler-time call evidence is useful, producing a separate cached JET result view instead of changing the main graph.
Start here if you want the shortest path from package setup to a working local analysis flow.
Learn what each Graph Explorer control does and how to read the graph once filters and reachability are applied.
Read how the backend, graph pipeline, engines, and frontend layers fit together.