Infrastructure Archaeology
See — live — what an error touches, and what a push will break.
Estates accumulate undocumented systems and forgotten dependencies. When something breaks or an update ships, nobody knows the blast radius. This tool discovers live systems, maps what depends on what, and shows the impact — not a replacement for human judgment, but a fast starting point for getting people back on track to where and how things went wrong.
$ arch scan 10.0.0.0/24
✔ discovered 214 hosts · 1,908 services (live)
! blast radius — if auth-svc fails:
38 services degrade · 2 SPOFs upstream
✔ start here → most-connected node highlighted
Representative output — not a live instance.
●Live discovery — unprivileged TCP scanning finds hosts and services across CIDRs; agents inventory local processes and sockets (Windows and Unix).
●Unified dependency graph — observations from many collectors merge into one typed graph with stable IDs, so re-scans converge instead of duplicating.
●Blast-radius analysis — computes exactly what breaks if a node fails, via reverse-reachability over the dependency edges.
●Single-point-of-failure detection — identifies the articulation points that partition your estate when removed.
●Interactive map — a force-directed graph with per-node blast radius, criticality rank, and a live event feed. Human-in-the-loop by design.
0
tests green
~8ms
SPOF detection, 5k-node graph
~16ms
criticality ranking, 5k nodes
live
SSE event stream
Rust · 9 cratesTokio · AxumHopcroft–Tarjan SPOFvanilla-JS force graphJSON / DOT export
🔒
Architecture and verified results shown. Source and binaries are private — this is a capability showcase, not a distribution.
Flying blind
on dependencies?
I build tools that map what you actually run and show the blast radius before you push. Tell me what you're afraid to touch.
Start a project →