Machine Payments Protocol (MPP) by Tempo and Stripe

# Fix list: Machine Payments Protocol (MPP)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mpp, the October 2026 research run, assessed 1 October 2026. Grade A, 81.1 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Machine Payments Protocol (MPP): work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Transparency & trust, 45 out of 100, up to 4.8 more on the total

Made of editorial 67, provenance 23.

Why it scored 45: Specs under CC0 and tooling and SDKs under MIT or Apache-2.0 (30). The draft's privacy section says servers must not require accounts and should not log credentials, but there's no privacy statement for mpp.dev or the directory, and the card path runs under Stripe's terms (15). The versioning section commits to a new scheme name for breaking changes, and the move from draft-ryan-httpauth-payment to draft-httpauth-payment is recorded on the datatracker per the 26 September check (14). No legal entity or governance body is named, and the authors work at Tempo Labs and Stripe (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Legal entity named: not found (0 of 20)
- Domain age: mpp.dev, registered 2024-07-13 (2 years) (7 of 15)
- Status page: not found (0 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 2. Security & auth, 79 out of 100, up to 3.7 more on the total

Why it scored 79: Payment proofs are single use, challenges can be bound with HMAC-SHA256 and to a digest of the request body, and the card path uses Stripe shared payment tokens with usage and expiry limits (26). Spend is capped by SPT limits and session deposits, while overall agent budgets are left to the client (14). The draft tells clients they MUST NOT trust the description and MUST check amount, recipient, currency and validity window themselves (13). Every paid response carries a Payment-Receipt header and settlement leaves an on-chain or Stripe record (13). Reports go to security@tempo.xyz with acknowledgement in 5 days, there's no bug bounty while Tempo is under audit, and mppx has published five advisories in 2026, one critical (13).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 3. Reliability, 85 out of 100, up to 3 more on the total

Why it scored 85: Graded as a protocol on reference implementations (30), public servers or processors (25), spec stability (25) and test vectors (20). Official SDKs in TypeScript (mppx), Python, Go, Rust and Ruby (30). There's no facilitator role, servers verify on chain or through Stripe, and the mpp.dev directory lists 137 services, a count the operators publish without a date (20). The core is an individual Internet-Draft at revision 01 (9 September 2026), not adopted by any IETF working group, while the scheme carries no wire version and is built to evolve by adding fields (15). The draft includes HMAC-SHA256 challenge-binding test vectors, and every SDK runs a shared conformance suite from tempoxyz/mpp-tools in CI, with cross-SDK smoke tests in Rust (20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 4. Schema & documentation, 89 out of 100, up to 1.8 more on the total

Why it scored 89: Each method draft defines a request schema and a credential schema, and the core defines the challenge, credential and Payment-Receipt formats, though we found no standalone JSON Schema files (18). llms.txt at mpp.dev with about 180 links (10). A 1,464-line draft in RFC 2119 language, including when not to return 402 and how Payment interacts with other auth schemes (20). Typed parameters per method and intent (13). Eleven error codes mapped to HTTP status, Retry-After guidance and worked examples of challenges and credentials (15). A versioning section and draft revisions, with SDK changelogs and a blog for changes (13).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Agent ergonomics, 91 out of 100, up to 1.5 more on the total

Why it scored 91: One 402 and one retry, and session intents let many small calls share one deposit (23). Charge and subscription intents, per-method sessions, client payment preferences and a discovery draft (17). Error codes such as payment-insufficient, payment-expired and invalid-challenge say what failed, and servers SHOULD send Retry-After (18). The draft requires single-use proofs, at most one settlement for concurrent requests with the same credential, and recommends an Idempotency-Key for non-idempotent methods (18). SDKs in five languages and a CLI (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 6. Payments & pricing, 94 out of 100, up to 0.8 more on the total

Why it scored 94: MPP is a machine payment protocol (40). No protocol fee. Tempo gas is capped near $0.0006 a transfer and Stripe charges 1.5 per cent on stablecoins and $0.15 per shared payment token per the 26 September check, while Stripe's page sets minimums of $0.50 for card tokens and 0.01 USDC for stablecoins (17). The spec is CC0 and the SDKs are open source (20). A buyer with a stablecoin wallet needs no account, but the card route needs a token from Link, optionally approved by a person (17).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 7. Maintenance & community, 95 out of 100, up to 0.4 more on the total

Why it scored 95: mppx 0.12.0 and mpp-rs 0.14.0 on 29 September 2026, and a spec commit on 30 September (30). mppx alone tagged six releases in September (20). 14 open issues on mppx, the newest filed on 29 September, many labelled friction (22). Official SDKs in five languages, though mpp-go's last tag is v0.2.0 from 1 July 2026 despite later commits (13). CI with dependency scanning, Dependabot, OpenSSF Scorecard in mppx and fuzzing in Rust (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 2026-03-26, three advisories in mppx eight days after launch, GHSA-8x4m-qw58-3pcx (critical, multiple payment bypass and griefing bugs), GHSA-mv9j-8jvg-j8mr (high, Tempo session close voucher bypass) and GHSA-8mhj-rffc-rcvw (moderate, Stripe credential replay). Fixed and published, so we deduct 2 (https://github.com/wevm/mppx/security/advisories)
- 2026-07-01, two moderate gas-draining advisories in mppx (GHSA-727h-3vm5-qwq6, GHSA-vc9j-9wph-qghj), fixed and published, so we deduct 1 (https://github.com/wevm/mppx/security/advisories)

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Whether any IETF working group will adopt the draft, and who owns the spec if not.
- unchecked: Stripe's current fees for MPP stablecoin and card payments, which the page we read doesn't state.
- How many of the 137 directory services take MPP directly rather than through a proxy run by someone else.
- Which findings in arxiv 2609.00060 apply to MPP and whether they're fixed.

## Weaknesses

- Individual Internet-Draft, not adopted by any IETF working group
- Five mppx advisories in 2026, one critical
- No legal entity or governance body named for the spec
- Stripe's minimums are $0.50 for card tokens and 0.01 USDC for stablecoins
- No bug bounty while Tempo is under audit

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Check amount, recipient and currency in the `request` parameter, never the description
- Send an `Idempotency-Key` when retrying a paid POST
- Use a session for many small calls to one server rather than a charge per call
- Set `max_amount` and `expires_at` on any card token
- Run a current mppx, releases before 0.4.11 had payment-bypass and session-voucher bugs

## What the review panel asked for

- Current Stripe fees
- State Stripe's MPP fees on one page

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.