# Fix list: Adobe Commerce (Magento) From Anchor Terminal's listing at https://www.anchorterminal.com/tools/adobe-commerce, the October 2026 research run, assessed 8 October 2026. Grade B, 63.9 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 Adobe Commerce (Magento): 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, 55 out of 100, up to 9 more on the total Why it scored 55: Graded with the hosted lines, because Adobe runs the cloud service and Adobe Commerce on Cloud. status.adobe.com lists Adobe Commerce with its components and regions (20). Its event feed shows eight Commerce events between 1 and 23 September 2026, three marked major. Catalogue Service, Live Search and the Admin panel had 2 hours 10 minutes of availability and performance trouble on 1 September, Live Search 45 minutes on 2 September, and Commerce Admin and provisioning on the cloud service 1 hour 39 minutes on 18 September. None names the REST or GraphQL endpoint itself, so 5 of 30 where several majors would score 0 (5). Input limits are published with numbers (20 items in a list input, 5,000 for asynchronous calls, 300 a page, GraphQL complexity 300 and depth 20), but no request rate limit was found (5). The server-to-server sample retries on 429 after `retry-after`, with no written guidance and no idempotency keys found (5). The Unified SLA's Actionability Addendum of 23 July 2026 covers Adobe Commerce. We didn't read the percentage (10). Generally available (10). 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total Why it scored 30: No x402, MPP or L402 in the API docs or the schemas (0). business.adobe.com timed out on every request on 8 October 2026, so the pricing page is unread and scored as absent. The cloud service licensing terms define an account as one opened under a sales order (0). Magento Open Source, the core with the same REST and GraphQL framework, is free under OSL 3.0 with no card (20). Its Composer install needs authentication keys from a Commerce Marketplace account, which a person creates. The public Git repository can be cloned with no account and the install guide points contributors to it, so half the points (10). 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. Security & auth, 60 out of 100, up to 7 more on the total Why it scored 60: The cloud service takes OAuth 2 tokens from Adobe IMS, by client credentials with the `commerce.accs` scope or by a user flow tied to an Admin role. A 2.4 store issues integrations whose requests are signed by OAuth 1.0a, admin tokens that last 4 hours and customer tokens that last 1 hour, and bare integration tokens as bearer tokens are off by default. No secret in a URL was found. Scopes stop at the product level, so finer limits come from roles (26). Each integration or admin role is limited to chosen ACL resources. No read-only switch or confirmation step for writes was found (12). The APIs return shopper-written text such as names, order comments and reviews, and no prompt-injection guidance was found (0). Action Logs record admin users' changes with user, IP address and date on Adobe Commerce, not on Magento Open Source. Whether integration calls are logged wasn't established (8). security.txt is valid to 30 July 2027 and names a public Intigriti programme and PSIRT, and security patches are tied to numbered bulletins. The compliance list didn't show its Commerce rows to our reader (14). 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. Agent ergonomics, 67 out of 100, up to 5.4 more on the total Why it scored 67: `?fields=` trims any REST response and GraphQL returns only the fields asked for. The surface is large, at 510 REST operations and 246 GraphQL queries and mutations on the cloud service (23). `searchCriteria` gives filters with comparison operators such as `eq`, `in` and `like`, sort orders, `pageSize` and `currentPage`, and GraphQL lists page the same way (20). REST errors are an HTTP code and a message with parameters. GraphQL answers 200 for business errors with a message and a category, and no stable list of codes was found (12). No idempotency keys found. Asynchronous and bulk endpoints return a bulk UUID whose status can be read back, and PUT by SKU updates in place (6). List calls need the long `searchCriteria` syntax, 2.4 stores need OAuth 1.0a signing, and the only official client library is `@adobe/aio-commerce-sdk` for JavaScript (6). 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, 82 out of 100, up to 2.9 more on the total Why it scored 82: Swagger 2.0 YAML files in the public docs repository for the cloud service (510 operations, version September 2026) and for the admin, customer and guest APIs of 2.4.6 to 2.4.9, plus GraphQL introspection JSON for each (25). Every page on developer.adobe.com is also served as Markdown at the same path with `.md`. No llms.txt there, and Experience League has one for the operations and admin guides (10). All 510 cloud service operations carry a description, with a median length of 39 characters and 147 under 30, and 239 of 246 GraphQL queries and mutations have one. None says when not to use a call (11). Definitions are typed and GraphQL uses enums, while `searchCriteria` parameters are plain strings and `custom_attributes` and `extension_attributes` are open (10). Tutorials with full requests and responses, a status code table, and 400, 401 and 500 responses in the spec, with no list of error codes (11). A schema per release, monthly cloud service release notes and release notes for each 2.4 version (15). 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, 75 out of 100, up to 2.2 more on the total Why it scored 75: The cloud service's latest production release is dated 7 October 2026 (30). Four cloud service releases since August 2026, in August, twice in September and in October. The 2.4 line last shipped on 12 May 2026 with 2.4.9 and security patches for 2.4.6 to 2.4.8 (20). The API docs repository merged pull requests on 8 October 2026. magento/magento2 has 1,230 open issues and we couldn't read how recent the replies are, because the GitHub API refused us for its rate limit (12). `@adobe/aio-commerce-sdk` is at 2.0.0 with alpha builds published on 8 October 2026. It is for App Builder and JavaScript only, and no Adobe Commerce server is in the official MCP registry (8). The docs repository carries tests and link checks. CI on the core repository wasn't read (5). 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. Transparency & trust, 90 out of 100, up to 0.9 more on the total Made of editorial 80, provenance 99. Why it scored 90: Magento Open Source is OSL 3.0. Adobe Commerce is proprietary under Adobe's General Terms (2025v1, 3 October 2025) and a product-specific document for each deployment, all public (22). The privacy policy of 24 October 2025 says Adobe is the processor for enterprise products and that the policy doesn't cover data its customers collect. The cloud service terms keep customer data for the licence term and allow 30 days to export after it ends. A data processing addendum is published, which we didn't read (22). The lifecycle policy gives three years of standard support and end dates for every release from 2.4.4 to 2.4.9, and says Cloud environments on unsupported versions lose traffic from 1 June 2027. No notice period for API changes on the cloud service was found (18). The sub-processor list, updated 28 September 2026, names seven for Adobe Commerce with countries or regions, among them Microsoft, Platform.sh, Fastly and New Relic (18). 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): - Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10) ## 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: the pricing page at business.adobe.com/products/commerce/pricing.html, which timed out on every request, so no price, trial or sandbox offer was read and Payments scores it as absent - unchecked: the security bulletins at helpx.adobe.com (APSB26-05 and APSB26-49 among them), which answered 403, so severities and any exploitation in the last 12 months weren't read and no deduction was considered - unchecked: which certifications Adobe lists for Adobe Commerce. The compliance list didn't show its product rows to our reader - unchecked: the uptime percentage and credits in the Unified SLA and its Actionability Addendum, and the text of the General Terms and the data processing addendum - unchecked: how quickly issues on magento/magento2 get a reply, and whether its CI passes. The GitHub API refused us for its rate limit - No request rate limit for the cloud service's REST or GraphQL APIs was found in the reviewed documentation - Whether calls made by an integration appear in Action Logs wasn't established - The sub-processor list names Microsoft for hosting, while the status page's registry lists Commerce Cloud regions on AWS, Azure and GCP ## Weaknesses - No request rate limit for the REST or GraphQL APIs was found in the reviewed documentation, and 429 handling appears only in a code sample - No idempotency keys were found, so a retried POST can create a second order or cart item - status.adobe.com lists three Commerce incidents marked major in September 2026, two of them longer than an hour - The cloud service drops the customer and guest REST APIs, so shopping flows there go through GraphQL only - No Adobe MCP server for Commerce was found, and the only official client library is a JavaScript package for App Builder ## 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. - Ask which deployment the store is. The cloud service answers at `https://.api.commerce.adobe.com//` with no `/rest` prefix, and other stores at `https:///rest//` - On the cloud service, request an IMS token with the `commerce.accs` scope and send the `Store` header to pick the store view - On a 2.4 store, sign requests with all four integration credentials by OAuth 1.0a. A bare integration token is refused unless the merchant has switched that on - Add `?fields=` to REST calls and set `searchCriteria[pageSize]`. Stores with input limits on cap a page at 300 items - Don't retry `placeOrder` or a POST blindly. Read the cart or order back first, because no idempotency key 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.