x402 by x402 Foundation (Linux Foundation)

# Fix list: x402

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/x402, the October 2026 research run, assessed 1 October 2026. Grade A, 79.7 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 x402: 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. Security & auth, 67 out of 100, up to 5.8 more on the total

Why it scored 67: Each payment is a signed authorisation for one amount, one recipient and one time window, not a reusable bearer secret, though the published attacks show the binding to the requested resource can fail (22). The upto scheme caps an amount, but spend budgets live outside the spec (10). The PaymentRequired body is untrusted input from the seller, and the spec doesn't tell clients to check amount and payTo against their own expectations (6). Settlement returns a transaction hash, so every payment has an on-chain record (13). SECURITY.md routes reports to Coinbase's HackerOne bug bounty, a high-severity advisory (GHSA-qr2g-p6q7-w82m) was fixed and published on 6 March 2026, and the spec has drawn two academic analyses this year, one validating five attacks on 12 May 2026 (16).

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.

## 2. Transparency & trust, 65 out of 100, up to 3.1 more on the total

Made of editorial 72, provenance 57.

Why it scored 65: Spec and SDKs under Apache-2.0 (30). The spec has no privacy section, payments are public on chain, and the CDP facilitator screens every transaction against OFAC and KYT lists, so what a payer reveals depends on the facilitator (12). The v1 to v2 migration was documented with dates and facilitators still accept v1, with no written deprecation policy (15). Governance is named, a technical steering committee of Coinbase, Cloudflare and Stripe and a charter dated 31 March 2026 under LF Projects, but security reports still go to Coinbase (15).

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):

- Domain age: x402.org, registered 2025-02-20 (1 year) (3 of 15)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## 3. Reliability, 87 out of 100, up to 2.6 more on the total

Why it scored 87: Graded as a protocol on reference implementations (30), public facilitators (25), spec stability (25) and test suites (20). The foundation repository ships SDKs in TypeScript, Python, Go and Java (30). The docs list 15 public facilitators from separate operators, among them Coinbase CDP, PayAI, Dexter, Polygon, Stellar and Fireblocks (25). The core is v2, stable since 9 December 2025, but schemes and networks are added most weeks, v1 headers are still accepted, and issue 3015 reports the x402.org testnet facilitator crashing on a malformed v2 request (20). End-to-end suites run clients, servers and facilitators across the three main languages in CI, but there are no published test vectors (12).

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. Agent ergonomics, 84 out of 100, up to 2.6 more on the total

Why it scored 84: One 402 and one retry, with base64 JSON in the PAYMENT-REQUIRED and PAYMENT-SIGNATURE headers (22). Exact, upto, auth-capture and batch-settlement schemes, and discovery through the Bazaar extension (16). Error codes name the failed check (amount mismatch, expired authorisation, wrong recipient, insufficient funds), and settlement_pending carries the transaction hash so a client can reconcile before retrying (17). EIP-3009 nonces and validity windows stop replays on EVM, the SVM libraries cache settlements against duplicates, and a payment_identifier extension exists, but the five published attacks include replay and binding failures (14). Official SDKs in four languages with fetch and axios wrappers (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.

## 5. Schema & documentation, 86 out of 100, up to 2.3 more on the total

Why it scored 86: The core spec defines PaymentRequired, PaymentPayload, SettlementResponse and VerifyResponse as JSON with field tables, and the facilitator interface (/verify, /settle, /supported), but ships no JSON Schema files (18). llms.txt at docs.x402.org pointing at .md pages (10). A 729-line core spec with the flow, per-scheme and per-network documents, and a template for new schemes (17). Fields typed with required flags and CAIP-2 network identifiers (13). JSON examples throughout and a list of standard error codes, including the non-terminal settlement_pending (14). Version history in the spec, per-package changelogs and a v1 to v2 migration guide (14).

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.

## 6. Maintenance & community, 93 out of 100, up to 0.6 more on the total

Why it scored 93: Releases on 30 September 2026, TypeScript and Go at 2.28.0 and Python at 2.25.0 (30). Eleven release trains between 10 July and 30 September 2026, roughly weekly (20). 223 open issues and heavy pull-request traffic, and the newest open issues date from late July and early August, which suggests newer reports get closed, though we couldn't see reply times (18). Current official SDKs in TypeScript, Python, Go and Java (15). CI per language, lint, package-lock checks and end-to-end tests (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.

## 7. Payments & pricing, 97 out of 100, up to 0.4 more on the total

Why it scored 97: x402 is a machine payment protocol (40). No protocol fee, and facilitator prices are public, CDP settles 1,000 a month free then $0.001 each per the 30 September check, and several facilitators charge nothing, but the FAQ tells mainnet users to hold ETH for gas while the exact scheme says the facilitator pays it (17). Free to implement under Apache-2.0, with a free testnet facilitator (20). A buyer needs only a funded wallet and no account (20).

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.

## Deductions

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

- 2026-03-06, GHSA-qr2g-p6q7-w82m (high). Facilitators processing Solana payments on @x402/svm before 2.6.0, Python x402 before 2.3.0 or Go before 2.5.0 were exposed. Keys and funds weren't affected, and the fix and advisory were public, so we deduct 2 (https://github.com/x402-foundation/x402/security/advisories/GHSA-qr2g-p6q7-w82m)
- 2026-05-12, five attacks on x402 validated on local chains, Base Sepolia and live endpoints, across authorisation, binding, replay and web handling, causing unpaid service or paid-but-denied outcomes. Some related fixes appear in the repository (origin binding for sign-in, SSRF in Bazaar), but we couldn't confirm all five are closed, so we deduct 1 (https://arxiv.org/abs/2605.11781)

## 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 all five attacks in arxiv 2605.11781 are fixed in current SDKs.
- Which of the 40 formal-analysis findings in arxiv 2609.00060 apply to x402 and whether any were disclosed to maintainers.
- Whether the x402 Foundation will take over vulnerability intake from Coinbase's HackerOne.
- The x402.org transaction counter has no date or method, so we didn't use it.

## Weaknesses

- Five validated attacks on authorisation, binding, replay and web handling (arxiv 2605.11781)
- A high-severity facilitator advisory on Solana handling in March 2026
- Spend budgets sit outside the spec, only the upto scheme caps an amount
- The FAQ and the exact-scheme spec disagree on who pays gas
- Security reports still go to Coinbase's HackerOne rather than a foundation channel

## 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.

- Decode PAYMENT-REQUIRED and check amount, asset and payTo against what you expected before signing
- Use the upto scheme when the final price isn't known, and cap it
- On settlement_pending, look up the returned transaction hash before paying again
- Use a production facilitator for Base mainnet, x402.org/facilitator is testnet only
- Give the agent its own wallet with a small balance, never a treasury key

## What the review panel asked for

- Fix gas wording
- Client budgets
- Reconcile the FAQ and exact-scheme gas rules
- Put client spend budgets in the spec

## 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.