A primary-source review of MPP, Payment Auth, Tempo and HTTP 402 for the week ending August 9, 2026.
Last week was quiet on integrations and busy on spec hardening. This week keeps the integration count at zero, but the spec work moved somewhere more interesting: three different settlement chains had their session or charge semantics change in the same seven days, and the SDK-facing documentation went through the largest consolidation pass I’ve tracked in this brief so far.
Three chains, three spec changes, one week
mpp-specs PR #258 adds a Hedera session intent, built around ERC-20 escrow secured with EIP-712 cumulative vouchers. mpp-specs PR #305 adds confidential transfer support to the Solana charge spec. mpp-specs PR #309 tightens Solana’s operator-signed session mode and adds negotiated idle-timeout handling, including server-initiated idle close.
None of these is a new service or a new integration — they’re changes to how sessions and charges behave on chains MPP already supports. But three separate chains getting substantive settlement-primitive changes in one week is a different kind of signal than the usual documentation churn: it reads as the spec authors working through chain-specific edge cases (confidentiality on Solana, escrow semantics on Hedera, session lifecycle on both) rather than adding surface-level features. That’s the kind of work that has to happen before a multi-chain protocol can be trusted with production volume on more than one rail.
mppx’s documentation got a large consolidation pass
Five merged pull requests synced the mppx documentation against the current implementation: a full main-branch snapshot sync covering discovery, transports, x402 requests and Tempo sessions; new request attestation documentation, including Web Bot Auth and Trusted Agent Protocol examples; a pass to make the relay integration guide easier to scan; a sync of replay claims and method hooks; and a relay announcement on the MPP blog showing developers how to delegate payment-method settlement.
The attestation docs are worth flagging on their own. Web Bot Auth and Trusted Agent Protocol are both about proving a request came from a legitimate, identifiable agent before payment logic runs — that’s the Payment Auth layer this brief tracks separately from the payment rail itself. Documenting it doesn’t mean it’s newly implemented; the PR is docs, not a spec change. But it’s a sign the project wants that surface to be legible to integrators now, not later.
One UI fix worth naming, and a pile that isn’t news
mpp PR #859 fixes service arrows in the UI to link to provider websites — small, but it’s the kind of fix that affects whether a prospective integrator can actually find the service behind a catalog entry. The rest of this week’s repository activity — code-highlight contrast, an operator-ligature rendering bug, marketing page header colors — is real work, and correctly not treated as news here. Holding that line is the point of separating “busy commit history” from “busy adoption story.”
A settlement-index challenge to the volume narrative
A Hacker News post, “Show HN: Merona – x402 settlement index”, claims raw x402 feeds overstate settlement volume by roughly 100x. I don’t have primary evidence beyond the post — no independent verification of Merona’s methodology, no corroborating data from a protocol-side source — so I’m naming it as a signal to watch, not a finding to report. But it’s exactly the kind of claim this brief exists to flag rather than let pass: x402 volume numbers get cited casually in the ecosystem, and an independent index alleging a two-order-of-magnitude gap between raw feeds and real settlement deserves scrutiny before it’s repeated as fact in either direction.
What I’m taking from a spec week, not an adoption week
Zero new services, zero new provider integrations, and zero new payment terms verified this week — the catalog didn’t move. What did move was underneath it: three chains’ session and charge semantics matured, the SDK documentation caught up to the implementation, and an outside party raised a real question about how much of the reported x402 activity is actually settling. None of that is adoption evidence, and I’m not going to call it that. It’s the protocol continuing to get more precise about what it supports, on more chains, while the catalog holds still. Next week’s honest question is still the one from two weeks ago: does anything already listed show a second week of real usage.
I built MPP Pulse to keep those layers separated instead of blurring them into one undifferentiated “things happened” feed. It’s an open-source AWS agent that collects primary evidence and drafts a cited weekly brief for human review.