ACK-Pay (Agent Commerce Kit)

# Fix list: ACK-Pay (Agent Commerce Kit)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/ack-pay, the October 2026 research run, assessed 9 October 2026. Grade D, 53.6 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 ACK-Pay (Agent Commerce Kit): 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. Reliability, 38 out of 100, up to 12.4 more on the total

Why it scored 38: Graded as a protocol on reference implementations (30), public servers or processors (25), spec stability (25) and test vectors (20). One reference SDK in TypeScript at 0.11.0, with a payments demo and example issuer and verifier services in the repository (15). No public Payment Service or Receipt Service was found. The docs use example.com hosts and the demo settles on Base Sepolia, Solana devnet or a simulated Stripe payment (0). The versioning page names 2025-05-04 as the current stable protocol version and says the date changes only on incompatible changes, but the specification is prose documentation without normative language and the SDK is below 1.0 (15). No test vectors are published. The ack-pay package has seven unit test files, and the Check workflow passed on the last five pushes to main on the page read (8).

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.

## 2. Agent ergonomics, 47 out of 100, up to 8.6 more on the total

Why it scored 47: A paid call takes a 402, a call to a Payment Service, a receipt and a retry, which is one party more than a direct 402 scheme. A receipt can be reused until it expires (15). A request can carry several payment options, there are server-initiated and client-initiated sequences and an optional `serviceCallback`, and no session or metering scheme is defined (12). No protocol error codes. The SDK throws two named error classes and the operations page asks implementers to write their own error taxonomy (5). The request `id` is there to stop replay and verifiers are told to reject reused receipts, with no idempotency key (8). One SDK, in TypeScript (7).

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.

## 3. Security & auth, 58 out of 100, up to 7.4 more on the total

Why it scored 58: Payment Requests are signed JWTs and receipts are Verifiable Credentials checked against a list of trusted issuer DIDs, with revocation through a status list. Open issue 222 reports that `expiresAt` is not enforced (22). A human oversight page describes approval before execution, left to each Payment Service, and the demo's policy guard is labelled illustrative (10). The docs tell implementers to isolate LLM interfaces from signing components and list four receipt checks, with no warning about the free-text `description` (8). Receipts are signed records that can carry references to policy, mandate and settlement evidence (11). SECURITY.md gives a reporting address. No bug bounty, no security.txt and no published advisory, although 0.11.0 fixed a fail-open revocation check (7).

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.

## 4. Schema & documentation, 66 out of 100, up to 5.5 more on the total

Why it scored 66: The SDK exports Valibot and Zod schemas for the Payment Request, the payment option and the receipt claim, with no JSON Schema or OpenAPI file (15). llms.txt lists 39 pages and each has a Markdown copy (10). Pages on roles, both sequences, the Payment Service and receipts explain when each flow applies, in descriptive prose without MUST or SHOULD rules (13). A field table gives type and required status for every request field, while `network` and `recipient` are free strings whose format varies by network (10). JSON examples of a request and a receipt are given, and no error responses are defined (7). Date-based protocol versions and a changelog per package, though the docs and SDK disagree in places. The docs allow a DID in `serviceCallback` and open issue 205 reports the schema rejects it (11).

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. Maintenance & community, 53 out of 100, up to 4.1 more on the total

Why it scored 53: 0.11.0 of every package was tagged on 4 August 2026, 66 days before the check (20). That is the only release in the last 90 days, and nine changesets wait on main (0). 16 open issues, the oldest from 10 August 2026, and 37 open pull requests. Outside contributions were merged on main in September and on 6 October 2026 (15). One official SDK, current on npm (8). CI on every push, Dependabot and a scheduled audit workflow (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.

## 6. Transparency & trust, 60 out of 100, up to 3.5 more on the total

Made of editorial 63, provenance 57.

Why it scored 60: Docs and code are MIT, copyright Catena Labs, Inc. (30). No privacy policy or terms were found for agentcommercekit.com. The protocol pages say receipt metadata must not hold secrets or customer data and that the Payment Service hides the server's settlement account from the client (10). The versioning page defines Draft, Current and Final revisions, and the 0.11.0 changelog lists each removed API with its replacement (13). The ack-pay package depends on four workspace packages and valibot, and no telemetry was found in its manifest. The docs site names no legal entity, which appears only in the repository's LICENCE file and package manifests (10).

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: agentcommercekit.com, registered 2025-05-02 (1 year) (3 of 15)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## 7. Payments & pricing, 77 out of 100, up to 2.9 more on the total

Why it scored 77: ACK-Pay is a payment protocol with a 402 request and a verifiable receipt, but it names no settlement method of its own and no public Payment Service was found to pay through (25). Scored with the self-hosted rule for the rest. No protocol fee, with any conversion fee left to the Payment Service to disclose (20). Docs and SDK are MIT with nothing to buy (20). The SDK installs from npm with no account, but a payment needs a Payment Service and a Receipt Service that somebody has to run (12).

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.

- 2026-08-04, @agentcommercekit/ack-pay 0.11.0 fixed a revocation check that treated any failure to fetch or parse the status list as not revoked, so `verifyPaymentReceipt` accepted revoked credentials. The changelog describes the flaw in full and the repository has no published advisory. Fixed and documented, so we deduct 2 (https://github.com/agentcommercekit/ack/blob/main/packages/ack-pay/CHANGELOG.md, https://github.com/agentcommercekit/ack/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 Payment Service or Receipt Service runs ACK-Pay in production. None is named in the pages read.
- unchecked: catenalabs.com. Its robots.txt request answered with a redirect that we did not follow, so the vendor's own site, blog and any hosted product were not read.
- unchecked: the client-initiated sequence, use cases, summary, demo and ACK-ID pages of the docs, which were not read to keep to the page budget.
- unchecked: the threads of the 16 open issues. Issue 222 on `expiresAt` was read by title only. In the source `createPaymentRequestToken` passes no expiry option to the JWT, which agrees with the title.
- Whether the rendered docs pages link terms or a privacy policy in a footer. The Markdown copies and llms.txt link none.
- Which protocol version follows 2025-05-04 and when. The roadmap lists x402 facilitation as planned research with no date.

## Weaknesses

- No public Payment Service or Receipt Service was found. The docs use example.com hosts and the demo pays on Base Sepolia, Solana devnet or a simulated Stripe flow
- The protocol defines no error codes, idempotency key or retry header. The operations page tells implementers to design their own
- One SDK, in TypeScript, at 0.11.0. One release in the last 90 days, with nine changesets waiting on main, including a receipt option check
- 0.11.0 fixed a revocation check that accepted revoked credentials when the status list was unreachable. No GitHub advisory was published
- Open issue 222 of 21 September 2026 reports that `expiresAt` on a Payment Request is never enforced
- No terms, privacy policy, status page or security.txt on agentcommercekit.com

## 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.

- Verify `paymentRequestToken` and read amount, recipient and currency from the signed payload, not from `description`
- Pass `trustedReceiptIssuers` to `verifyPaymentReceipt`. Without a trusted issuer list any DID can sign a receipt
- Check amount, recipient and receipt reuse yourself. The SDK's claim verifier checks only the shape of the receipt
- Use `@agentcommercekit/ack-pay` 0.11.0 or later, because earlier releases treated an unreachable revocation list as not revoked
- Plan for a Payment Service and a Receipt Service you run or choose. The protocol names the roles and no public one was found

## 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.