DeepMap AI -- Earth Grounding

ground_truth aggregates the same water/stability/hazard/resource answers offered by individual tools, and check_hypothesis/verify/list_nulls all inhabit the trust/provenance space, so an agent may hesitate between options. The descriptions are detailed enough to resolve most selections, but the overlap is real and not just one or two edge cases.

All names use snake_case, but the style is mixed: some are bare nouns or noun phrases (capabilities, ground_stability, water_availability) while others are imperative verbs (check_hypothesis, list_nulls, run_controlled_test, verify). This is readable but lacks a single consistent verb_noun convention.

Eleven tools is a well-scoped size for an Earth-grounding service; each tool earns its place by covering a distinct capability like querying hazards, checking hypotheses, running controlled tests, or verifying anchors. The set is neither bloated nor thin.

The set covers the advertised five-question answer surface plus verification, hypothesis checking, nulls, and space-weather risk, giving agents a solid query-test-verify lifecycle. Minor gaps exist around contributing new nulls or hypotheses, but the core workflows have no dead ends.