# State of the Agent Economy · 2026-W41 > Week of Oct 5, 2026 · 4 sources cited ## Headline numbers - **Indexed**: 11,112 - **Live**: 1,159 - **Earning (30d)**: 1,747 - **Gross USD (30d)**: $910,736.46 - **Gini**: 0.994 ## Briefing This is an **early-week snapshot**, not a completed ISO week: as of October 5, 2026, W41 has just begun, so its figures should not be read as a full-week run rate or a like-for-like comparison with W40. The clearest signal is that measured agent availability and observed earning wallets are rising, while payment activity remains highly concentrated and most payment-protocol records still rely on declarations rather than live runtime evidence. ## Measurement: more indexed and live agents, but not all signals mean the same thing Agenstry’s index reached **11,112**, up **945 (+9.3%)** from W40. Live endpoints rose to **1,159**, up **216 (+22.9%)**, and the earning-wallet count reached **1,747**, up **218 (+14.3%)**. The separate 30-day growth measure shows **3,133** new agents, up **399 (+14.6%)** from the previous period. These measures describe different stages and should not be collapsed into a single adoption rate. The earning figure counts wallets with observed inbound USDC to declared x402 pay-to addresses, across the wallets scanned so far; it is not an ecosystem-wide revenue or participation total. Likewise, the weekly payment-protocol cohort currently shows **34** records for W41, all x402, but the cohort is only a partial-week observation. ## Payments: broad protocol listings, narrow realized revenue The 30-day gross total is **$910,736.46**, but the distribution is sharply skewed: the top agent accounts for **69.5%** of measured gross, the top 10 for **92.1%**, and the Gini coefficient is **0.994**. The median is **$0.41**, compared with a **$101.52** 95th percentile. The HHI stands at **4,909**, down **794 (-13.9%)** from W40, while Gini moved from **0.995** to **0.994**. The data therefore suggest slightly lower concentration by those two measures, but not broad-based earnings: the median remains small relative to the tail. Agenstry lists **4,721** x402 payment-protocol records, against **123** for MPP, **59** for A2A-x402, and **34** for AP2. Those are protocol records, not settled-payment volumes. Of **4,943** labeled entries, **1,958 (39.6%)** have self-declared-live evidence—an endpoint’s 402 challenge or a served x402 manifest. The dataset itself notes that runtime challenges are authoritative, so static cards and directory listings should not be treated as proof that a paid endpoint currently works. ## Spec & protocol: A2A gets a command-line front door The most concrete protocol-facing launch surfaced this week is the A2A project’s October 1 announcement of its official CLI. It is intended to let users and tools discover agents, send messages, and manage interactions from a terminal, broadening how A2A endpoints can be exercised in development and operations. [A2A Protocol announcement](https://a2a-protocol.org/latest/blog/) That is an implementation and usability step, not a new protocol specification. The published A2A roadmap still emphasizes v1.1 robustness, bidirectional streaming, and validation tooling. [A2A roadmap](https://github.com/a2aproject/A2A/blob/main/docs/roadmap.md) For operators, the practical question is whether easier invocation is accompanied by repeatable conformance testing and useful runtime observability. ## Payments infrastructure: x402 hardening continues The x402 Python SDK’s September 29 release added Arc mainnet and testnet support and addressed a dependency issue affecting SVM integrations. The changelog also documents recent fixes to payment middleware, including decoded-path route matching—an important reminder that implementation security remains central to payment-gated services. [x402 Python SDK changelog](https://github.com/x402-foundation/x402/blob/main/python/x402/CHANGELOG.md) MCP did not show a newer specification release in the sources reviewed: its latest stable specification remains the July 28, 2026 revision, whose stated changes include a stateless core and authorization hardening. [MCP release history](https://github.com/modelcontextprotocol/modelcontextprotocol/releases) The distinction matters for buyers: protocol adoption, declared payment support, and verified successful settlement are separate milestones. ## What to watch next week The next useful check is whether W41’s partial-week x402 cohort expands and whether more listed payment endpoints pass live runtime probes. A rising record count alone would be less informative than evidence that declared services are reachable, correctly enforce payment, and produce settlement activity beyond the most concentrated earners. ## Sources 1. [A2A Protocol announcement](https://a2a-protocol.org/latest/blog/) 2. [A2A roadmap](https://github.com/a2aproject/A2A/blob/main/docs/roadmap.md) 3. [x402 Python SDK changelog](https://github.com/x402-foundation/x402/blob/main/python/x402/CHANGELOG.md) 4. [MCP release history](https://github.com/modelcontextprotocol/modelcontextprotocol/releases) --- Read the live (HTML) version with charts + live numbers: https://agenstry.com/reports/state-of-agent-economy Archive: https://agenstry.com/reports/state-of-agent-economy/archive