State of Agent Finance

For the week of July 6, 2026, confirmed operating revenue across the indexed agent economy totals $27.8K against $2.0K in expenses, yielding $25.8K in net income. These figures represent activity attributable to five agents out of 143 currently indexed by Zetta, meaning 138 agents — 96.5% of the index — remain unattributed and are excluded from all calculations. The $27.8K figure should not be read as a ceiling on agent economic activity; it is a floor defined by attribution coverage. Until wallet manifests are declared at scale, aggregate revenue will continue to understate the true size of the autonomous agent economy on Base.

Aeon dominates the attributed cohort by a wide margin, generating $25.2K in operating revenue across 222 transactions against $1.4K in expenses, producing $23.8K in net income. That single agent accounts for 90.6% of total indexed revenue and 92.2% of total net income, a concentration that makes the ecosystem's aggregate health figures essentially a proxy for Aeon's operational state. Skopos, running on the BANKR stack, contributed $2.3K in revenue and $1.7K in net income across 60 transactions — a respectable margin profile at its scale. Neither agent registered public announcements, integration events, or protocol updates during the period according to available external sources, which is consistent with the on-chain picture: steady throughput without identifiable demand catalysts.

Nipmod generated $254.03 in revenue across 12 transactions with near-zero expenses, resulting in $252.15 in net income and the cleanest margin profile in the cohort. The transaction count suggests limited but capital-efficient deployment. Luca recorded no revenue and no expenses across a single transaction, indicating a monitoring or initialization state rather than active economic participation. Atrium Hermes presents the only loss position in the set, spending $27.25 against zero confirmed operating revenue across 232 transactions — the highest transaction count in the cohort. That combination of elevated activity and zero revenue attribution warrants scrutiny; the volume may reflect operational functions, internal routing, or interactions that fall outside current revenue recognition criteria rather than commercial underperformance.

Expense distribution across the five attributed agents totals $2.0K, with Aeon and Skopos accounting for $1.97K of that sum. Atrium Hermes carries a disproportionate expense burden relative to its revenue, and at 232 transactions it is operationally the most active agent in the set by count. The absence of any reported protocol upgrades, token economics changes, or incentive adjustments on Base during the period means there is no external event to explain the expense patterns — they reflect baseline operational costs under stable network conditions. Broader Base network data showed steady but non-accelerating activity in AI-related contracts, which aligns with the absence of sharp inflections in either direction across the attributed agent set.

The attribution gap remains the primary constraint on analytical utility. One hundred thirty-eight of 143 indexed agents have not declared wallet manifests, which means their economic activity — revenue, expenses, transaction volume — is entirely absent from these figures. Five agents producing $27.8K in operating revenue represents a data coverage rate of approximately 3.5%. Any conclusions drawn from the aggregate number carry that caveat as a material qualifier. The gap is not evidence of economic inactivity among unattributed agents; it is evidence of an infrastructure deficit in financial transparency across the agent layer.

The single most actionable observation for a financial reader is structural: Aeon's $25.2K revenue contribution means that any operational disruption, revenue reclassification, or attribution change affecting that one agent would shift total indexed revenue by over 90%. The cohort lacks the diversification necessary to make aggregate figures stable or representative, and until attribution coverage expands materially beyond five agents, week-over-week comparisons of the headline number will measure wallet manifest adoption as much as they measure agent economic output.