Each tool targets a distinct operation: leis_buscar lists matching norms, leis_facetas aggregates counts, leis_obter retrieves a single norm's text, leis_cobertura reports index/coverage diagnostics, and leis_autoria handles authorship. The shared filter vocabulary between buscar and facetas is explicitly signposted, and their outputs (items vs. counts) are clearly different, so an agent can select confidently.

All names share the same 'leis_' namespace prefix and are lowercase snake-style, which makes the set predictable. The only wrinkle is a mixed verb/noun pattern (buscar, obter are verbs; facetas, cobertura, autoria are nouns), but this still reads naturally and doesn't impede selection.

Five tools is a tight, well-scoped surface for a local legal-norm archive: one search, one fetch, one aggregation, one diagnostics, one authorship. Nothing appears redundant and no obvious capability is bolted on unnecessarily.

The read-only lifecycle for the domain is well covered: discovery (buscar), retrieval (obter), aggregation (facetas), corpus health (cobertura), and provenance (autoria). Minor gaps exist — no dedicated version-history/comparison tool despite versao_texto being exposed, and no explicit export/raw-text operation — but agents can work around these via obter.