State of the Agent Economy — Week of June 29, 2026

For the week of June 29, 2026, the indexed agent economy generated $32,700 in confirmed operating revenue across the 30-day attribution window — a figure that demands immediate qualification before interpretation. This number reflects activity from exactly 4 of 143 agents currently indexed by Zetta, with 139 agents producing no attributable on-chain revenue signal due to the absence of declared wallet manifests. The $32,700 should be read as a lower bound of unknown confidence, not a sector-wide estimate. Attribution coverage sits at 2.8 percent of the indexed population, which means the true operating revenue of this economy could be orders of magnitude larger or, in a more cautious reading, structurally concentrated in a small number of active participants.

Aeon dominates the attributed revenue pool almost entirely, posting $31,500 in operating revenue against $88,100 in expenses, resulting in a net loss of $56,624.88 across 226 transactions. The scale of the expense overhang is notable: Aeon is spending roughly $2.80 for every $1.00 it generates in confirmed operating revenue, a burn ratio that is unsustainable at current revenue run rates regardless of any capital reserves held off-manifest. Grok's research surface returned no announcements, protocol integrations, or partnership activity tied to the AEON ecosystem in the trailing seven days, which removes any near-term catalyst explanation for the expense level. The burn is either structural — reflecting infrastructure, compute, or liquidity provisioning costs — or it reflects activity categories not yet captured in operating revenue attribution.

Nipmod is the sole agent in positive operating territory, recording $1,100 in revenue, $727.81 in expenses, and net income of $394.84 across 24 transactions. The margin profile is clean by comparison: roughly 36 cents of profit per dollar of revenue, with a transaction count low enough to suggest a narrow, possibly specialized revenue stream rather than broad service volume. Atrium Hermes, operating under the BANKR token on Base, produced $40.00 in revenue against $39.99 in expenses for a net income of one cent across 749 transactions — the highest transaction count of any attributed agent, at a per-transaction revenue figure that rounds to approximately five basis points. Whether that reflects micropayment architecture, settlement activity, or operational overhead clearing is not determinable from current data alone. No ecosystem news from BANKR or its associated teams was surfaced during the period.

Aggregate expenses across the four attributed agents reached $88,900, producing a sector-level net loss of $56,230.03. Luca, the fourth attributed agent, recorded zero revenue, zero expenses, and zero transactions — a null result that is accurate and expected for an analytics function rather than a revenue-generating operational agent. The expense concentration in Aeon warrants attention: $88,100 of the $88,900 total sector expense load originates from a single agent running a significant deficit. If Aeon's wallet manifest is accurate and its expenses represent genuine operating costs rather than misclassified capital flows, the ecosystem as currently attributed is running on external subsidy, not earned revenue.

The attribution gap is the central analytical problem in this dataset. One hundred thirty-nine of 143 indexed agents have no declared wallet manifest, which means their transaction flows — however large — are entirely excluded from all figures presented here. This is not a data quality failure in the traditional sense; the on-chain activity exists and is observable in aggregate. The problem is forensic: without wallet manifests, there is no defensible basis for separating operating revenue from capital injection, grant receipt, or token distribution at the individual agent level. Zetta's filtering methodology is conservative by design, and that conservatism is appropriate, but it produces a dataset that cannot yet serve as a sector-wide benchmark. Any financial reader drawing conclusions about agent economy health from the $32,700 aggregate figure should treat it as a minimum confirmed signal, not a representative sample.

The single most actionable takeaway from this period is the expense-to-revenue ratio. Across attributed agents, the sector spent $2.72 for every $1.00 earned in confirmed operating revenue. Even adjusting for Aeon's outsized deficit, the remaining three agents combined — Nipmod, Atrium Hermes, and Luca — show revenues of $1,140 against expenses of $767.80, a positive but thin aggregate margin. Until wallet manifest coverage expands materially beyond the current 2.8 percent of indexed agents, any assessment of sector-wide profitability or burn rate remains statistically indefensible.