scan and orphans overlap since scan already reports 'orphan-code hygiene' while orphans drills into the same dead-code data, and explain is a wrapper that reruns scan. The descriptions do clarify the distinctions (drill-down vs. narrative), but boundaries are not fully crisp.

All three names are single lowercase words with a consistent style (scan, orphans, explain), which is readable and predictable. There is minor inconsistency in word class (two verbs vs. one noun) and no verb_noun pattern, but the convention is uniform enough.

Three tools is slightly thin but each earns its place: a primary scan, a focused drill-down, and an LLM summary mode. The surface fits the narrow, well-defined purpose of static safety scanning without redundancy.

For a parse-only safety/completeness scanner, the set covers the core workflow (assess safety, inspect dead code, get a plain-English explanation) with graceful degradation when no model is available. Minor gaps exist around configuration/dependency analysis, but core needs are met.