A primary-source review of MPP, Payment Auth, Tempo and HTTP 402 for the week ending August 2, 2026.
Last week was an expansion week: a first-party integration, four new services, new supporting infrastructure for spend authorization and wallets. This week didn’t repeat that. No new payment terms, integrations or provider announcements verified in the primary record for the seven days ending August 2. That’s worth saying plainly rather than padding around — the interesting things this week happened one layer down, or one layer over.
Refund safety got more precise, not more capable
The most substantive change was a specification refinement in mpp-specs PR #299, clarifying Near Intents refund and settlement recovery. The change binds refund destination validation before a quote request can proceed, aligning the spec with hardening work already tracked in the implementation repository.
This isn’t a new feature. It’s the kind of change that makes an existing feature harder to get wrong. Refund handling is exactly the surface where an underspecified protocol creates real financial risk — a refund sent to an unvalidated destination is a fund-loss bug, not an edge case. Tightening that ahead of wider Near Intents usage is a sign of a spec being hardened deliberately rather than reactively, which is the more reassuring order of operations.
Onboarding got easier, in three small pieces
Three documentation additions target the same problem — the distance between “MPP exists” and “I have working code”: a Privy wallet integration guide, a relay integration guide, and an MPPX integration carousel for discoverability. None of these ship a new capability. They reduce the friction of using capabilities that already exist, which matters more than it sounds like once a protocol has more than a handful of implementers to onboard.
The rest of this week’s repository activity was dependency bumps, layout fixes and marketing-style polish — real work, correctly not treated as news. Holding that line matters as a design goal for this brief: a busy commit history isn’t the same claim as a busy adoption story.
Tempo’s bigger news this week was about Tempo, not MPP
Two posts on the Tempo blog are worth separating out because they’re a different kind of signal: an announced expansion of Tempo’s onchain yield offering with BlackRock’s BUIDL fund, and a research post on liquidity mobility in markets.
Neither is an MPP protocol change. But Tempo is MPP’s primary payment rail, and an institutional yield product built on top of it is a statement about who Tempo wants underwriting that rail long-term. It’s a company-level move, not a spec-level one — I’m flagging it rather than folding it into “MPP developments,” because conflating the two is exactly the kind of category error this brief exists to avoid.
One community signal worth watching, not verifying yet
A Hacker News post, “Show HN: Tilde Pay – Give your AI agent a bank account to pay for things”, describes an agent that sells its own API access over x402. I don’t have primary evidence beyond the post itself — no verified integration, no transaction record — so I’m naming it as a signal to watch rather than a development to report. It’s the kind of independent, agent-initiated payment experiment that the broader x402/MPP interoperability question is ultimately about, and it’s worth another look if it resurfaces with more evidence behind it.
What I’m taking from a quiet integration week
Zero new verified integrations doesn’t mean zero movement — it means the movement was in safety hardening, developer experience and the business layer underneath the protocol, rather than in the catalog. That’s a normal week for infrastructure that’s past its launch phase and hasn’t yet had its second adoption wave. Next week’s honest question is the same as last week’s: does Exa, or any of the four database APIs, 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.