discovery-graph

Build an infrastructure inventory by capturing raw host, network, service, storage, and repository discovery data, then translating the evidence into a durable SQLite graph with explicit entities, relations, and rerunnable discovery runs.

Store discovery evidence in SQLite by running a full discovery sweep that captures raw host and service data, normalizes it into a shared discovery model, and persists a replayable graph with a durable run_id for inactive-removal and evidence-backed audits.

Automating infrastructure discovery requires raw-capture and persistence helpers like linux_collect.py, capture_run.py, discovery_store.py, and normalize_minimum.py to gather evidence, normalize it, and persist a concrete graph in SQLite.

Inspect machines and repos by executing discovery commands that gather raw evidence, then apply bootstrap normalization tooling to translate the captured data into a shared SQLite discovery model with entities and relations.

A SQLite discovery model replaces ad hoc shell output by translating raw discovery evidence into durable entities, relations, and rerunnable discovery runs, leaving behind explicit evidence-backed graph data for reproducible audits.

A full discovery sweep supports inactive-removal semantics by capturing fresh evidence, normalizing it into the SQLite discovery model, and updating the graph to reflect active and inactive infrastructure relations accurately.