UXMachine

Each tool has a fairly distinct role: medir starts a measurement, estado polls its status, memoria reads historical site memory, and tunel sets up a local tunnel. The only mild overlap is between estado and memoria, since both surface observed results, but the descriptions clarify that estado is per-measurement status while memoria is prior-site memory.

All names are single lowercase Spanish words, which is stylistically consistent, but there is no predictable verb_noun pattern and the parts of speech are mixed (verb 'medir' vs nouns 'estado', 'memoria', 'tunel'). Readable but not a systematic convention.

Four tools map cleanly onto the async measurement workflow (start, poll, prior-memory, tunnel), so each earns its place. It is slightly lean, with no separate tool for listing or canceling measurements, but not thin enough to feel incomplete.

The core lifecycle—launch a measurement, poll until done, review prior site memory, and enable local-server measurement—is covered. Minor gaps exist around managing measurements (listing history, canceling, credit/account status) that an agent could work around but would occasionally miss.