No user-submitted related servers found.
Scored across 14 tools
Most tools target clearly distinct concerns (citation grounding, evidence confidence, invariance, trust scoring, provenance, audit chain, decision grading), and each description explicitly names when to use a sibling tool instead. The one real soft spot is check_payout_invariance vs check_mutation_invariance, where the former is described as a special case of the latter, plus some conceptual overlap between check_provenance_claims and check_claims_registry, though the descriptions disambiguate both.
All names follow a uniform snake_case verb_noun pattern (check_grounding, corroborate_evidence, append_audit_entry, verify_audit_chain, grade_decision, etc.). Verb choice varies (check/score/assess/grade/compute/scaffold) to reflect genuinely different actions, but the structural convention is consistent throughout.
14 tools sits comfortably in the healthy 3-15 range, and each has a distinct, non-trivial purpose rather than being padding. The set is broad (citation checks, invariance tests, trust scoring, audit chains, decision grading, and two scaffold helpers), so it functions more as a curated honesty toolkit than a single tightly-scoped domain, but no tool feels redundant.
Coverage is broad for an 'honesty/integrity' toolkit, with natural pairs such as append_audit_entry/verify_audit_chain and score_trust_identified/assess_anonymous_authenticity covering both sides of each concern. Minor gaps exist (e.g. no dedicated truth-of-evidence checker, and grounding explicitly defers semantic entailment to an external kit), but the surface covers the stated intent without obvious dead ends.
