Spree Commerce

# Fix list: Spree Commerce

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/spree-commerce, the October 2026 research run, assessed 9 October 2026. Grade BB, 70.4 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 Spree Commerce: 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, 65 out of 100, up to 6.1 more on the total

Why it scored 65: Publishable keys for the Store API and secret keys with `read_*` and `write_*` scopes per resource for the Admin API, shown once, revocable, with last-used dates. Keys travel in headers and have no expiry (28). `read_all` gives a read-only key, the CLI mints only read-only keys by itself, and a key cannot create a key with wider scopes. No confirmation step for destructive calls was found (15). The API returns merchant and shopper text, and no prompt-injection guidance was found in the reviewed documentation (5). The audit trail is Enterprise only. The open-source edition has last-used dates on keys and a webhook delivery log (5). `SECURITY.md` with a security mailbox, a 72-hour acknowledgement target, advisories published on GitHub, Dependabot and a weekly audit workflow. No bounty, no security.txt and no certification (12).

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. Payments & pricing, 55 out of 100, up to 5.6 more on the total

Why it scored 55: Read with the self-hosted rule. No x402, MPP or L402 (0). The Community Edition is free and its scope is public, while the Enterprise Edition is listed as custom pricing (15). Nothing to pay and no card for the open-source edition (20). An agent can install a store with `npx create-spree-app@latest`, which has an option for every prompt and prints the API keys, and the CLI mints a read-only key in a local project (20). The hosted Sandbox is free but needs a browser sign-up and is for evaluation only.

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.

## 3. Reliability, 82 out of 100, up to 3.6 more on the total

Why it scored 82: Graded as software the owner runs. Installs with `npx create-spree-app@latest` or the `spree` gem, with Node, Docker, Ruby 3.2 or later and Rails 8.1 stated (20). A public Tests workflow runs on Postgres and MySQL, and most of the 25 latest runs listed on 9 October 2026 completed successfully, with two failures on pull-request branches (22). 95 open issues against 1,798 commits since 10 July 2026, many of them roadmap items. Reply times were not read (18). The release policy is a shifted semantic versioning in which minor releases may break, with an upgrade guide per minor and `CHANGELOG.md` pointing to GitHub releases (9). 5.6.1 is stable, but the default docs describe the 6.0 release candidate and `@spree/admin-sdk` is 0.1.0 on npm (13).

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

Made of editorial 75, provenance 45.

Why it scored 60: BSD 3-Clause for the backend, MIT for the npm packages and CC BY 4.0 for the docs. Enterprise Edition code is under a separate commercial licence and is not in the repository (28). The software runs on the operator's servers, so store data stays with the operator. Vendo's only privacy notice covers its websites and the hosted Sandbox, with lead data kept up to 24 months and recipients named by category only (20). The release policy pairs breaking changes with deprecation notices in the previous release and lists supported versions with end dates, but the two published copies of that list disagree (12). No telemetry was found in the CLI or installer source, and an open issue titled Telemetry and Update Check plans it (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):

- Endpoint on the vendor's domain:  is not on spreecommerce.org (0 of 15)
- Terms of service: not found (0 of 10)
- Privacy policy: not found (0 of 10)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## 5. Schema & documentation, 85 out of 100, up to 2.4 more on the total

Why it scored 85: OpenAPI 3.0.3 files for the Store API (81 paths, 99 operations) and Admin API (259 paths, 417 operations) are public (25). llms.txt and a Markdown copy of each page (10). Every Admin operation and 91 of 99 Store operations carry a description, which says what the call does and rarely when not to use it (13). Typed parameters with required fields and 48 enums in the Store file, though `fields`, `expand` and Ransack `q[...]` filters are free strings (11). Error codes are tabled with JSON examples, and the Store file documents 401, 404 and 422 responses but not 429 (13). The path carries `/api/v3`, docs are split into v5.x and v6.x, and a dated changelog has an RSS feed, but the default version documents a release candidate (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.

## 6. Maintenance & community, 81 out of 100, up to 1.7 more on the total

Why it scored 81: The stable server release is 5.6.1 of 28 July 2026, 73 days before the check (20). `@spree/cli` 3.2.2 and the 6.0.0.rc1 tag are dated 8 October 2026 and are not counted for recency. More than three releases in 90 days, among them 5.6.0, 5.6.1, 5.5.4, 5.4.4 and five 6.0 pre-releases (20). 95 open issues and 86 open pull requests, the newest issues opened on 8 and 9 October, and 1,798 commits since 10 July. Reply times were not read (17). Current official TypeScript SDKs and CLI on npm (15). Dependabot weekly for gems and Actions, a weekly `pnpm audit` workflow and CI on every pull request (9).

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. Agent ergonomics, 90 out of 100, up to 1.6 more on the total

Why it scored 90: `fields` trims a response, `expand` adds relations that are left out by default, and `limit` caps a page (23). `page`, `limit` up to 100, `sort` and Ransack filters on list calls, with a `meta` block of counts (20). Errors carry a machine-readable `code`, a `message` and per-field `details`, with 34 documented codes (19). `Idempotency-Key` on eight cart, checkout and payment calls with a 24-hour cache, optimistic locking on carts, and SDK retries with backoff. Guest cart creation is not covered (18). Few required parameters and three TypeScript SDKs plus a CLI, with no official client in a second language (10).

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.

## Deductions

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

- 2026-07-20: six security advisories were published for `spree_api` and related gems between 8 January and 20 July 2026, four rated High, among them guest address and order disclosure without authentication (GHSA-87fh-rc96-6fr6, GHSA-p6pv-q7rc-g4h9, 5 February) and a guest cart takeover on `PATCH /api/v3/store/carts/:id/associate` (GHSA-4825-p4xm-pcf2, 20 July, fixed in 5.4.4 and 5.5.4). All are published with patched versions, so the deduction is reduced (https://github.com/spree/spree/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.

- unchecked: spree.sh and console.spree.sh were not read, because the Sandbox terms forbid scraping without Vendo's consent. Sandbox limits and the sign-up flow are known only from the terms and the CLI docs
- unchecked: reply times on GitHub issues and pull requests
- unchecked: the patched versions of four of the six 2026 advisories. Two advisory pages were read in full
- unchecked: the tool definitions of the docs MCP server at https://spreecommerce.org/docs/mcp. The docs say it has two tools, search and read
- Docs pages other than those fetched as Markdown were read from the repository's `docs` folder at commit 4ca24b8, which is the source of the published site
- No terms or privacy document governs the self-hosted software. The only published terms and privacy notice cover the hosted Sandbox, so both provenance fields are left out and the BSD 3-Clause licence stands in
- Which supported-version table is current. `SECURITY.md` gives 5.x until 26 March 2027 and 4.10 until 6 September 2026, and the docs give 5.0 until 26 March 2028 and 4.10 until 6 September 2027
- The 5.5 to 5.6 upgrade guide refers to a 5.6.2 release, and no 5.6.2 tag or gem was found

## Weaknesses

- Six GitHub security advisories between 8 January and 20 July 2026, four rated High, most of them access-control flaws on guest carts, orders and addresses
- The default docs and the main-branch OpenAPI files describe 6.0, a release candidate dated 8 October 2026. The stable release is 5.6.1 of 28 July 2026
- The audit trail is in the paid Enterprise Edition only, and Enterprise prices are not published
- Supported-version end dates differ between `SECURITY.md` (5.x until 26 March 2027) and the docs security policy (5.0 until 26 March 2028)
- The official MCP server searches and reads the docs only. Store and admin MCP servers are open roadmap issues

## 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.

- Send a publishable key in `X-Spree-Api-Key` on every Store API call. Add `X-Spree-Token` with the cart token for guest carts, or a customer JWT as a Bearer token
- Send `Idempotency-Key` with the cart token or a customer JWT on cart, payment and completion calls. A request carrying only the publishable key is never cached
- Ask for `fields` and `expand` explicitly. Relations are left out by default and `limit` stops at 100
- Use a secret key with the narrowest scopes for the Admin API. `read_all` gives a read-only key, and the CLI's auto-minted local key is read-only
- Read the docs under `/docs/v5` for a 5.6 store. The default docs describe 6.0, which renames shipments to fulfilments and splits carts from orders

## 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.