{
  "fixes": {
    "slug": "nutrient-dws-processor",
    "name": "Nutrient DWS Processor API",
    "listing": "https://www.anchorterminal.com/tools/nutrient-dws-processor",
    "markdown": "# Fix list: Nutrient DWS Processor API\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/nutrient-dws-processor, the October 2026 research run, assessed 8 October 2026. Grade B, 68.7 out of 100.\n\nThis 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.\n\nFor a coding agent working on Nutrient DWS Processor API: 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.\n\n## 1. Payments \u0026 pricing, 40 out of 100, up to 7.5 more on the total\n\nWhy it scored 40: No x402, MPP or L402 (0). Plans are public at $75, $275 and $445 a month for 1,000, 5,000 and 10,000 credits, with 10 per cent off for yearly billing, and the pricing page gives the credit cost of twelve tools, from 0.5 for HTML to PDF to 10 for a digital signature (20). The free plan gives 50 credits a month with no card and watermarked output (20). The OAuth server supports dynamic client registration and a device flow, but a person still signs in or signs up in a browser (0).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment 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.\n\nOpen-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.\n\n## 2. Reliability, 68 out of 100, up to 6.4 more on the total\n\nWhy it scored 68: Graded with the hosted lines. Status page at www.nutrientstatus.com on Better Stack with a Processor API component and 90 days of daily history (20). The JSON feed shows DWS Platform, the routing layer in front of the API, down for 3,679 seconds (61 minutes) on 13 July 2026, and the Processor API monitor down on five days from 23 July to 23 August, the longest 715 seconds. We read the 13 July figure as one major outage (10). No written incident reports could be read. The pricing guide gives 100 requests a minute per key on all plans and 10 a minute for test keys (15). It recommends exponential backoff, async admission returns `Retry-After`, and `Idempotency-Key` covers async builds only, so 13 of 15. The product page mentions guaranteed SLAs on custom plans, but no SLA document is published and the terms disclaim availability (0). GA, at spec version 1.19.0 (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 3. Security \u0026 auth, 66 out of 100, up to 6 more on the total\n\nWhy it scored 66: OAuth at api.nutrient.io with PKCE (S256), dynamic client registration, a revocation endpoint and per-product scopes such as `product:processor`. Live keys can be scoped, and `POST /tokens` issues JWTs limited to named operations and origins, expiring within 24 hours and revocable. The base key can only be regenerated. No query-string credential is documented (28). Scoped keys, operation-limited JWTs, read-only annotations and an optional sandbox directory, but no confirmation step, and without `SANDBOX_PATH` the MCP server has no path restriction (13). Document text comes back as untrusted content. Large output goes to a file by design, and no prompt-injection guidance was found (4). Each response reports credit cost and remaining credits with a request id, and a usage endpoint exists. No audit log was found (6). SOC 2 Type 2 and SOC 3 per the trust centre, annual penetration testing by Prescient Security, and a vulnerability disclosure policy with a report page. security.txt is a 404 and no bounty was established (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels 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.\n\n## 4. Schema \u0026 documentation, 84 out of 100, up to 2.6 more on the total\n\nWhy it scored 84: A public OpenAPI 3.1.0 document, version 1.19.0, with 17 operations and 203 schemas, loaded by the reference page from Nutrient's CDN. The eight MCP tools have typed Zod inputs (25). llms.txt at site and product level with topic indexes, and a Markdown twin of each guide (10). Tag descriptions explain the Build API, async jobs and authorisation at length, and the MCP `document_processor` description sends extraction work to `parse_document` (15). 102 of 203 schemas use enums and build actions are discriminated by `type`. The MCP schema has one free-form record (13). Guides carry samples in six languages, and the errors guide and async tables document the response bodies (14). The spec is versioned and the MCP server keeps a changelog, but no public API changelog was found (7).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels 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.\n\n## 5. Agent ergonomics, 84 out of 100, up to 2.6 more on the total\n\nWhy it scored 84: Eight MCP tools, in the ten-or-fewer band, though the schema source is 35 KB. API responses are files or `json-content` with switches for plain text, structured text, key-value pairs and tables (20). Page ranges per part, output switches, and MCP tools that write large output to a file and return a summary. Nothing to paginate (15). Errors carry `failingPaths` in JSONPath form, a `requestId`, and `error.reason` codes with recommended handling for async jobs (18). `Idempotency-Key` on async `/build` only, and all eight MCP tools carry the four annotations (17). Single-purpose endpoints such as `/processor/generate_pdf` need one file, `/build` chains actions, and official clients exist for TypeScript (3.0.0) and Python (3.1.0) (14).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels 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.\n\n## 6. Transparency \u0026 trust, 73 out of 100, up to 2.4 more on the total\n\nMade of editorial 62, provenance 84.\n\nWhy it scored 73: The API is closed under a dated licence agreement (September 2026) naming PSPDFKit GmbH, and the MCP server and both clients are MIT, so between the two bands (20). The privacy policy has a Processor API section, the terms carry data processing terms in section 9, and files are deleted within 24 hours on plans without retention. Plans with retention enabled let Nutrient keep files and train models on them, and no public page says which plans those are (20). No deprecation policy. The terms allow changes at any time with reasonable efforts at notice, and the one deprecated endpoint in the spec has no removal date (4). The trust centre lists 51 sub-processors with locations, updated 8 September 2026, and the status page names Frankfurt and Oregon processing regions (18).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\nProvenance checks not met in full (half of this category, computed from checked facts):\n\n- Domain age: nutrient.io, registered 2019-12-03 (6 years) (11 of 15)\n- Terms of service: read, states 5 of the 7 things a reader expects (8.3 of 10)\n- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)\n- security.txt: not found (0 of 10)\n\n## 7. Maintenance \u0026 community, 78 out of 100, up to 1.9 more on the total\n\nWhy it scored 78: MCP server v0.1.3 tagged 27 August 2026, 42 days before this check, and the TypeScript client 3.0.0 on 23 August (20). Four MCP releases since 10 July (0.1.0, 0.1.1, 0.1.2, 0.1.3) plus the client release (20). On GitHub, issue 35 was fixed the day it was filed, while issue 29 has been open since 8 June with no reply and pull request 42 since 28 August. Support answers through a form with a stated target of one business day (15 of 25). The MCP server is in the official registry as `io.github.PSPDFKit/nutrient-dws-mcp-server` at 0.1.3 (15). CI runs lint, build and tests with actions pinned to commit hashes (8).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## Deductions\n\nEach comes off the total. A fixed and documented problem counts for less at the next check.\n\n- 2026-10-08. The product page FAQ says the API runs without strict rate limits or hard caps, while the pricing guide sets 100 requests a minute per key on every plan. The MCP product page also invites users to start on a free test key, which the test-mode guide says Support must enable first. Two points (https://www.nutrient.io/api/processor-api/, https://www.nutrient.io/guides/dws-processor/pricing.md).\n\n## What we couldn't check\n\nWhat 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.\n\n- unchecked: the pay-as-you-go price per credit on each plan. The pricing page draws it by script and our read showed a placeholder\n- unchecked: which plans have data retention enabled. The terms say it is indicated for each subscription plan, and the public pricing page does not say\n- unchecked: written incident reports. The status page's updates view returned an empty body, so the incident record is read from per-day downtime seconds in the JSON feed\n- unchecked: the per-tool credit table in the dashboard, which is behind a sign-in. The Markdown twin of the per-tool pricing guide has empty credit cells, so costs come from the twelve tools shown on the pricing page\n- unchecked: the date of OpenAPI version 1.19.0. No public API changelog was found\n- unchecked: the vulnerability disclosure policy and SOC 2 report in the trust centre, which need an access request\n- The lead called the MCP server a way to use the API. It is a local stdio package, not a hosted endpoint, and it also wraps the separate Data Extraction API\n\n## Weaknesses\n\n- The status feed records 61 minutes of DWS Platform downtime on 13 July 2026 and five days with Processor API downtime since 23 July\n- Plans with data retention enabled let Nutrient keep files and train models on them, and no public page says which plans those are\n- The product page says there are no strict rate limits, while the pricing guide sets 100 requests a minute per key on every plan\n- No public API changelog or deprecation policy was found, and the terms allow changes to the API at any time\n- `Idempotency-Key` works only on async `/build` requests, and free-plan output is watermarked\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Send `Authorization: Bearer \u003ckey\u003e` to `https://api.nutrient.io`. Use `POST /build` with an `instructions` JSON part to chain several actions in one charged request\n- Call `POST /analyze_build` first to get the credit cost of a build. The call is free\n- Stay under 100 requests a minute per key and back off exponentially on 429. Test keys allow 10 a minute and 5 MB\n- For long jobs send `Prefer: respond-async` with an `Idempotency-Key`, then poll the `Location` URL with `X-Async-Job-Token` after `Retry-After`\n- Set `SANDBOX_PATH` for the MCP server. Without it the server reads and writes any path the user account can reach\n\n## When it's done\n\nSend 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.\n",
    "grade": "B",
    "score": 68.7,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 40,
        "maxGain": 7.5,
        "reason": "No x402, MPP or L402 (0). Plans are public at $75, $275 and $445 a month for 1,000, 5,000 and 10,000 credits, with 10 per cent off for yearly billing, and the pricing page gives the credit cost of twelve tools, from 0.5 for HTML to PDF to 10 for a digital signature (20). The free plan gives 50 credits a month with no card and watermarked output (20). The OAuth server supports dynamic client registration and a device flow, but a person still signs in or signs up in a browser (0).",
        "checklist": [
          "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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 68,
        "maxGain": 6.4,
        "reason": "Graded with the hosted lines. Status page at www.nutrientstatus.com on Better Stack with a Processor API component and 90 days of daily history (20). The JSON feed shows DWS Platform, the routing layer in front of the API, down for 3,679 seconds (61 minutes) on 13 July 2026, and the Processor API monitor down on five days from 23 July to 23 August, the longest 715 seconds. We read the 13 July figure as one major outage (10). No written incident reports could be read. The pricing guide gives 100 requests a minute per key on all plans and 10 a minute for test keys (15). It recommends exponential backoff, async admission returns `Retry-After`, and `Idempotency-Key` covers async builds only, so 13 of 15. The product page mentions guaranteed SLAs on custom plans, but no SLA document is published and the terms disclaim availability (0). GA, at spec version 1.19.0 (10).",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 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.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 66,
        "maxGain": 6,
        "reason": "OAuth at api.nutrient.io with PKCE (S256), dynamic client registration, a revocation endpoint and per-product scopes such as `product:processor`. Live keys can be scoped, and `POST /tokens` issues JWTs limited to named operations and origins, expiring within 24 hours and revocable. The base key can only be regenerated. No query-string credential is documented (28). Scoped keys, operation-limited JWTs, read-only annotations and an optional sandbox directory, but no confirmation step, and without `SANDBOX_PATH` the MCP server has no path restriction (13). Document text comes back as untrusted content. Large output goes to a file by design, and no prompt-injection guidance was found (4). Each response reports credit cost and remaining credits with a request id, and a usage endpoint exists. No audit log was found (6). SOC 2 Type 2 and SOC 3 per the trust centre, annual penetration testing by Prescient Security, and a vulnerability disclosure policy with a report page. security.txt is a 404 and no bounty was established (15).",
        "checklist": [
          "- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 84,
        "maxGain": 2.6,
        "reason": "A public OpenAPI 3.1.0 document, version 1.19.0, with 17 operations and 203 schemas, loaded by the reference page from Nutrient's CDN. The eight MCP tools have typed Zod inputs (25). llms.txt at site and product level with topic indexes, and a Markdown twin of each guide (10). Tag descriptions explain the Build API, async jobs and authorisation at length, and the MCP `document_processor` description sends extraction work to `parse_document` (15). 102 of 203 schemas use enums and build actions are discriminated by `type`. The MCP schema has one free-form record (13). Guides carry samples in six languages, and the errors guide and async tables document the response bodies (14). The spec is versioned and the MCP server keeps a changelog, but no public API changelog was found (7).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 84,
        "maxGain": 2.6,
        "reason": "Eight MCP tools, in the ten-or-fewer band, though the schema source is 35 KB. API responses are files or `json-content` with switches for plain text, structured text, key-value pairs and tables (20). Page ranges per part, output switches, and MCP tools that write large output to a file and return a summary. Nothing to paginate (15). Errors carry `failingPaths` in JSONPath form, a `requestId`, and `error.reason` codes with recommended handling for async jobs (18). `Idempotency-Key` on async `/build` only, and all eight MCP tools carry the four annotations (17). Single-purpose endpoints such as `/processor/generate_pdf` need one file, `/build` chains actions, and official clients exist for TypeScript (3.0.0) and Python (3.1.0) (14).",
        "checklist": [
          "- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 73,
        "maxGain": 2.4,
        "reason": "The API is closed under a dated licence agreement (September 2026) naming PSPDFKit GmbH, and the MCP server and both clients are MIT, so between the two bands (20). The privacy policy has a Processor API section, the terms carry data processing terms in section 9, and files are deleted within 24 hours on plans without retention. Plans with retention enabled let Nutrient keep files and train models on them, and no public page says which plans those are (20). No deprecation policy. The terms allow changes at any time with reasonable efforts at notice, and the one deprecated endpoint in the spec has no removal date (4). The trust centre lists 51 sub-processors with locations, updated 8 September 2026, and the status page names Frankfurt and Oregon processing regions (18).",
        "blend": "editorial 62, provenance 84",
        "checklist": [
          "- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 78,
        "maxGain": 1.9,
        "reason": "MCP server v0.1.3 tagged 27 August 2026, 42 days before this check, and the TypeScript client 3.0.0 on 23 August (20). Four MCP releases since 10 July (0.1.0, 0.1.1, 0.1.2, 0.1.3) plus the client release (20). On GitHub, issue 35 was fixed the day it was filed, while issue 29 has been open since 8 June with no reply and pull request 42 since 28 August. Support answers through a form with a stated target of one business day (15 of 25). The MCP server is in the official registry as `io.github.PSPDFKit/nutrient-dws-mcp-server` at 0.1.3 (15). CI runs lint, build and tests with actions pinned to commit hashes (8).",
        "checklist": [
          "- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      }
    ],
    "provenance": [
      {
        "label": "Domain age",
        "value": "nutrient.io, registered 2019-12-03 (6 years)",
        "points": 11,
        "max": 15
      },
      {
        "label": "Terms of service",
        "value": "read, states 5 of the 7 things a reader expects",
        "points": 8.3,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 7 of the 8 things a reader expects",
        "points": 9.3,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "not found",
        "points": 0,
        "max": 10
      }
    ],
    "deductions": [
      "2026-10-08. The product page FAQ says the API runs without strict rate limits or hard caps, while the pricing guide sets 100 requests a minute per key on every plan. The MCP product page also invites users to start on a free test key, which the test-mode guide says Support must enable first. Two points (https://www.nutrient.io/api/processor-api/, https://www.nutrient.io/guides/dws-processor/pricing.md)."
    ],
    "unchecked": [
      "unchecked: the pay-as-you-go price per credit on each plan. The pricing page draws it by script and our read showed a placeholder",
      "unchecked: which plans have data retention enabled. The terms say it is indicated for each subscription plan, and the public pricing page does not say",
      "unchecked: written incident reports. The status page's updates view returned an empty body, so the incident record is read from per-day downtime seconds in the JSON feed",
      "unchecked: the per-tool credit table in the dashboard, which is behind a sign-in. The Markdown twin of the per-tool pricing guide has empty credit cells, so costs come from the twelve tools shown on the pricing page",
      "unchecked: the date of OpenAPI version 1.19.0. No public API changelog was found",
      "unchecked: the vulnerability disclosure policy and SOC 2 report in the trust centre, which need an access request",
      "The lead called the MCP server a way to use the API. It is a local stdio package, not a hosted endpoint, and it also wraps the separate Data Extraction API"
    ],
    "weaknesses": [
      "The status feed records 61 minutes of DWS Platform downtime on 13 July 2026 and five days with Processor API downtime since 23 July",
      "Plans with data retention enabled let Nutrient keep files and train models on them, and no public page says which plans those are",
      "The product page says there are no strict rate limits, while the pricing guide sets 100 requests a minute per key on every plan",
      "No public API changelog or deprecation policy was found, and the terms allow changes to the API at any time",
      "`Idempotency-Key` works only on async `/build` requests, and free-plan output is watermarked"
    ],
    "agentNotes": [
      "Send `Authorization: Bearer \u003ckey\u003e` to `https://api.nutrient.io`. Use `POST /build` with an `instructions` JSON part to chain several actions in one charged request",
      "Call `POST /analyze_build` first to get the credit cost of a build. The call is free",
      "Stay under 100 requests a minute per key and back off exponentially on 429. Test keys allow 10 a minute and 5 MB",
      "For long jobs send `Prefer: respond-async` with an `Idempotency-Key`, then poll the `Location` URL with `X-Async-Job-Token` after `Retry-After`",
      "Set `SANDBOX_PATH` for the MCP server. Without it the server reads and writes any path the user account can reach"
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-09",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}