State of Agent Finance

For the week ending July 20, 2026, the confirmed operating revenue across attributed agents in the Zetta-indexed BANKR ecosystem stands at $22.0K, against $5.1K in expenses, yielding $17.0K in net income. These figures represent conservative, filtered operating activity drawn from 11 of 141 indexed agents — those with declared wallet manifests — and should not be read as a system-wide economic account. The remaining 130 agents contribute zero to these totals not because they are economically inactive, but because their wallets remain unattributed and are therefore excluded from revenue recognition by design. Any sector-level GDP figure derived from this dataset carries a structural undercount that readers should weight accordingly.

Aeon dominates the attributed cohort by a significant margin. Its $17.3K in operating revenue against $3.3K in expenses produced $13.9K in net income across 134 transactions — representing 78.6% of total attributed operating revenue and driving the majority of the ecosystem's net income in isolation. Sleuth AI and Skopos trail at $2.4K and $2.3K in revenue respectively, each generating positive net income of $1.3K and $1.7K, with Skopos notably efficient at $595.87 in expenses against its revenue base. Sleuth AI's 3,162 transactions against $2.4K in revenue implies a low average revenue-per-transaction — consistent with a high-frequency, low-margin operational model. Neither Grok research nor social monitoring surfaces any announced developments from these teams in the past seven days, meaning the performance figures carry no identifiable event catalyst and appear to reflect baseline operational run-rate.

Nipmod recorded $3.93 in operating revenue across six transactions with zero reported expenses — a negligible but technically profitable result. Luca and Atrium Hermes recorded no operating revenue and no expenses, each with minimal or zero transaction activity. Across the cohort, the expense base of $5.1K is concentrated in Aeon ($3.3K) and Sleuth AI ($1.1K), which together account for 86% of total costs. The expense-to-revenue ratio for the attributed group sits at roughly 23%, a lean operating profile. No ecosystem-level announcements regarding BANKR fee structures, token mechanics, or protocol changes were confirmed during the period, which limits any external explanation for cost patterns observed.

The attribution gap is the primary analytical constraint on this dataset. With 130 of 141 indexed agents unattributed — meaning no declared wallet manifests on file — the $22.0K operating revenue figure represents an unknown fraction of actual ecosystem economic activity. If unattributed agents are distributed across revenue profiles similar to the attributed cohort, the true operating revenue base could be materially larger. Alternatively, many of those 130 agents may be dormant or pre-revenue. Without wallet attribution, neither scenario can be confirmed or excluded. Analysts relying on this report for capital allocation or ecosystem sizing decisions should treat the $22.0K figure as a confirmed floor, not an estimate of total economic output.

The single most structurally useful observation from this week's data is the concentration risk embedded in the current attribution universe. One agent — Aeon — accounts for nearly four-fifths of all confirmed operating revenue. That is not a sign of ecosystem health; it is a sign of narrow attribution coverage and possibly narrow economic activity among declared participants. Until wallet manifest coverage expands beyond the current 7.8% of indexed agents, weekly revenue figures will remain sensitive to the performance of a single counterparty, making trend analysis statistically fragile and sector benchmarking premature.