Shopware

# Fix list: Shopware

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/shopware, the October 2026 research run, assessed 8 October 2026. Grade BB, 71.4 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 Shopware: 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. Payments & pricing, 50 out of 100, up to 6.3 more on the total

Why it scored 50: No x402, MPP or L402 (0). The paid plans have public starting prices, Rise from €600 and Evolve from €2,400 a month, with the real price set by GMV and Beyond on request (10). The MIT Community Edition is free with no card (20). An agent's owner can install it with Shopware CLI and Docker, and the install guide says no Shopware account is needed. A paid plan or SaaS shop needs a browser signup or sales (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.

## 2. Security & auth, 73 out of 100, up to 4.7 more on the total

Why it scored 73: Admin API by OAuth 2.0 client credentials from a revocable integration with an ACL role and 10 minute tokens, or the integration's key and secret in headers on MCP. No secret in a URL was found in the docs. No OAuth scopes beyond ACL roles, and key rotation wasn't found (25). ACL checks on every MCP call, allowlists per integration and per user, a global allowed_tools switch and dry run by default on write tools. shopware-media-upload has no dry run and the Store API MCP endpoint has no allowlist (17). The MCP best practices page has a section on prompt injection through order notes, names and product text and advises read-only integrations for such data (11). An integration records when it was last used, and no audit log of API or MCP calls was found (4). Bug bounty through the security reporting form per SECURITY.md, ISO/IEC 27001:2022 on the trust centre, and advisories published on GitHub. The security.txt file expired on 31 December 2025 (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.

## 3. Agent ergonomics, 78 out of 100, up to 3.6 more on the total

Why it scored 78: A fresh MCP session advertises three discovery tools and loads the rest by toolset. `includes` selects fields, and MCP results over 100 KB come back as a resource reference (25). page, limit, total-count-mode, filters, sorting and aggregations in one criteria object (20). MCP errors are a message written to say what to do next, and the APIs return JSON:API errors with codes. A 429 on MCP has no Retry-After header (16). No idempotency keys found. Write tools default to dryRun=true, but no MCP tool carries readOnlyHint or destructiveHint in the source (8). Few required parameters and sensible defaults (limit 25, maxResults 3). One official API client, @shopware/api-client for TypeScript, and none found in a second language (9).

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.

## 4. Reliability, 83 out of 100, up to 3.4 more on the total

Why it scored 83: Graded with the self-hosted package lines, because each store's APIs run on the merchant's own server or SaaS shop. Composer packages on Packagist with PHP 8.2 to 8.5 stated in composer.json (20). Nightly workflow on trunk, the last 10 scheduled runs all passing (25). 1,169 open issues, with bugs opened on 8 October 2026 already answered the same day and a triage labeller in the workflows (10). Semantic versioning with a written backward compatibility promise and UPGRADE and RELEASE_INFO files, but 6.7.14.0 changed what tools/list returns on the experimental Store API MCP endpoint in a minor release (13). 6.7 is stable, while the MCP server is experimental until 6.8 (15). Shopware SaaS has a status page at status.shopware.com, which lists a 36 minute outage of storefront and administration on 6 August 2026 and 1 hour 49 minutes of raised errors on 17 August. That isn't scored here.

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.

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

Why it scored 85: OpenAPI 3 for both APIs, as JSON in the repository, on a hosted Stoplight reference and from each instance at /_info/openapi3.json in dev mode. MCP tools have typed inputs (25). llms.txt with every docs page as Markdown (10). 109 of 116 Store API operations and 105 of 108 Admin API operations in the repository schema carry a description, and the MCP tool reference says when to use search, read or aggregate (16). Types and required fields are set, but the MCP tools take criteria, payload, aggregations and ids as JSON-encoded strings, and Admin API entity routes accept open criteria objects (9). Request examples in the guides, a JSON:API error schema and a fixed success and error envelope for MCP tools (12). No API version in the path since 2020, so the contract follows the product version, with release notes on GitHub and RELEASE_INFO files (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. Transparency & trust, 76 out of 100, up to 2.1 more on the total

Made of editorial 80, provenance 71.

Why it scored 76: MIT licence for the core (30). A self-hosted store keeps its data on the merchant's server. The privacy page, updated 18 August 2026, names shopware AG and links a data processing agreement, but no retention periods or sub-processors were found on it (18). A backward compatibility promise, `@deprecated` markers that name the major version of removal, and releases.json with an end date for security fixes on every version (18). A telemetry page lists what Shopware CLI, the Deployment Helper and the web installer send, by unencrypted UDP to a server in Frankfurt, with DO_NOT_TRACK as the opt-out. The core has a usage data module with consent that we didn't read (14).

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 shopware.com (0 of 15)
- Terms of service: read, states 5 of the 7 things a reader expects, and has 2 clauses that cost points (4.3 of 10)
- Privacy policy: published, but our reader couldn't read it (7 of 10)
- security.txt: published but past its Expires date (5 of 10)

## 7. Maintenance & community, 87 out of 100, up to 1.1 more on the total

Why it scored 87: 6.7.15.1 tagged on 2 October 2026 (30). Seven releases between 19 August and 2 October 2026, from 6.7.13.1 to 6.7.15.1 (20). Issues opened on 8 October 2026 had same-day replies, against 1,169 open issues and 322 open pull requests (17). @shopware/api-client 1.7.0 was published on 6 October 2026. The MCP server is built in and has no entry in the official registry, where the only Shopware server is a third party's (10). Nightly, integration, acceptance, static analysis and npm audit workflows in the repository (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.

- GitHub lists 20 security advisories for shopware/shopware published between 19 May and 16 September 2026, four of them critical (an app script sandbox escape, stored SQL injection through app manifests, admin account takeover by host-header poisoning and a webhook permission bypass), plus a pre-authentication SQL injection in the Store API (GHSA-p37c-pm9p-7vm5, CVSS 8.6, published 25 August 2026, fixed in 6.7.13.1 and 6.6.10.23). All were disclosed in public with fixed versions, so the deduction is 5 of a possible 15 (https://github.com/shopware/shopware/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: the security reporting page at www.shopware.com/en/contact/security-reporting/ answered 503, so the bug bounty's scope and rewards weren't read
- unchecked: the data processing agreement at www.shopware.com/en/privacy/dpa and any sub-processor list
- unchecked: the core's usage data module and what consent it asks for
- unchecked: the exact GitHub star count. The repository page showed 3.4k and the API refused us for its rate limit
- unchecked: whether an official API client exists in a second language
- The pricing page shows the plan prices in euros. US dollar prices for Rise and Evolve weren't in the page as fetched
- Whether the Store API MCP endpoint gains cart and checkout tools when it leaves experimental status in 6.8

## Weaknesses

- The MCP server is marked experimental until 6.8, and 6.7.14.0 changed what tools/list returns on the Store API endpoint
- 20 advisories published between 19 May and 16 September 2026, four critical, including a pre-authentication SQL injection in the Store API
- MCP tools carry no readOnlyHint or destructiveHint annotations, and criteria and payloads travel as JSON-encoded strings
- A 429 from the MCP endpoints carries the wait time in the body, with no Retry-After header
- The Store API MCP endpoint ships one domain tool, has no allowlist, and the security.txt file expired on 31 December 2025

## 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 the merchant for an integration without --admin, tied to an ACL role and an MCP allowlist. Send sw-access-key and sw-secret-access-key headers to /api/_mcp
- Call shopware-tool-search first, then shopware-toolset-enable, and keep the Mcp-Session-Id header. A fresh session lists only three tools
- Pass dryRun=false to commit a write. shopware-media-upload has no dry run and uploads at once
- For shopping, call the Store API over HTTP with the sales channel's sw-access-key and keep the sw-context-token. The Store API MCP endpoint has no cart tools in core
- Send `includes` in search criteria to cut response size, and read the 429 body for the wait time

## 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.