{
  "fixes": {
    "slug": "azure-translator",
    "name": "Azure Translator",
    "listing": "https://www.anchorterminal.com/tools/azure-translator",
    "markdown": "# Fix list: Azure Translator\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/azure-translator, the October 2026 research run, assessed 1 October 2026. Grade BB, 72.2 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 Azure Translator: 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, 20 out of 100, up to 10 more on the total\n\nWhy it scored 20: No x402, MPP or L402 (0). Per-million-character prices published without a login, readable through the Azure Retail Prices API since the pricing page needs JavaScript (20). F0 gives 2 million characters a month, but an Azure subscription needs a card (0). A person 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. Security \u0026 auth, 75 out of 100, up to 4.4 more on the total\n\nWhy it scored 75: Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Resource keys, 10-minute bearer tokens, or Microsoft Entra ID with the Cognitive Services User role and managed identities (30), less 10 because the authentication reference documents passing the key in a `Subscription-Key` query parameter (20). The data privacy page says text translation doesn't store customer data, which leaves nothing to train on, but we didn't find an explicit training statement there (15 of 20). Text isn't stored and documents are hard-deleted after processing. LLM translation runs through your own Foundry deployment, whose retention terms we didn't check (12 of 15). Azure Monitor and the activity log record resource actions, per-request logging not confirmed (10 of 15). MSRC disclosure policy, Azure bounty programme, SOC 2 and ISO 27001 reports. The microsoft.com security.txt passed its Expires date on 2026-09-23 (18 of 20).\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## 3. Reliability, 83 out of 100, up to 3.4 more on the total\n\nWhy it scored 83: Azure status page with post-incident reviews (20). No review in the last 90 days names Translator. One on 29 September 2026 covers intermittent failures for Azure OpenAI, Foundry and Cognitive Services in Sweden Central from 10:03 to 15:58 UTC, which may touch Translator resources and LLM deployments there, so we count it as minor. The public page lists only broad incidents (20). Limits published with numbers, 1,000 texts and 50,000 characters a request and hourly caps from 2 million characters on F0 to 200 million on S4 (15). 429 has three documented sub-codes and the limits page asks for load spread across the hour, but there's no Retry-After or backoff guidance for 429. Retry advice exists only for 408 and 503. Text translation has no side effects, so a retry is safe (8 of 15). The old Cognitive Services SLA address now redirects to Microsoft's Online Services SLA, a download we didn't open (10). API 2026-06-06 is GA (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## 4. Agent ergonomics, 82 out of 100, up to 2.9 more on the total\n\nWhy it scored 82: API reading of the checklist. Responses hold one translation per target with optional alignment and sentence lengths, and up to 1,000 texts go in one NMT call, but there's no field selection (20 of 25). Several targets per call, a `scope` filter on the language list, and `top` and `skip` paging on document jobs (18 of 20). Six-digit error codes such as 400036 (invalid target language) and 403001 (free quota exceeded) say what to fix (20). Text translation is stateless, but there's no 429 retry guidance and no idempotency key on document batches (12 of 20). A key, a region header and `api-version` on every call, with SDKs in Python, JavaScript, .NET and Java (12 of 15).\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## 5. Schema \u0026 documentation, 83 out of 100, up to 2.8 more on the total\n\nWhy it scored 83: OpenAPI documents for text translation 3.0 and 2026-06-06 are public in Azure/azure-rest-api-specs, with examples on each operation (25). No llms.txt found on learn.microsoft.com (0). Operation descriptions state purpose, and the overview says to pick NMT or an LLM by quality, cost and scenario, but not when to avoid either (15 of 20). Enums for `profanityAction`, `textType`, `tone` and `gender`, required fields marked, but the LLM `deploymentName` is a free string (13 of 15). Examples in the spec and a status-code page with six-digit error codes and their meanings (15). Dated `api-version` values and a What's New page (15).\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## 6. Maintenance \u0026 community, 70 out of 100, up to 2.6 more on the total\n\nWhy it scored 70: Read as a closed service. Document translation SDKs went GA in August 2026 (.NET 3.0.0 and JavaScript 1.0.0), after Python and Java 2.0.0 in July, and the text API 2026-06-06 in June (20, within 90 days). Four dated SDK releases in July and August (20). Public What's New page, Microsoft Q\u0026A and GitHub issue trackers on the SDK repositories, which we didn't read (10 of 15). Official SDKs current, `azure-ai-translation-text` 2.0.0 on 29 May 2026 for Python 3.9 to 3.13 (15). The SDKs live in the azure-sdk monorepos, CI not checked (5 of 10).\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## 7. Transparency \u0026 trust, 80 out of 100, up to 1.8 more on the total\n\nMade of editorial 65, provenance 95.\n\nWhy it scored 80: Closed service under Microsoft's product terms, SDKs under MIT (15). The data privacy page and privacy statement agree on no storage of text and hard deletion of documents, but say nothing on training, and the LLM path falls under separate Foundry terms (22 of 30). Microsoft's lifecycle policy applies, but the What's New page gives no retirement date for v3.0 even though 2026-06-06 breaks its request shape (10 of 20). Regional resources and endpoints for data residency and a public Microsoft sub-processor list (18 of 20).\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- security.txt: published but past its Expires date (5 of 10)\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- We didn't open the Online Services SLA document. The Cognitive Services SLA address redirects to it, which we took as coverage.\n- unchecked: retention and abuse-monitoring terms for LLM translation through a Foundry deployment.\n- The August 2026 SDK releases are dated by month only, so `lastRelease` stays at 2026-06-06.\n\n## Weaknesses\n\n- The 2026-06-06 schema breaks v3.0 clients and drops BreakSentence and dictionary lookups from the spec\n- LLM translation needs a Foundry deployment and bills tokens on the Azure OpenAI meter\n- Tone and gender controls only on the LLM option, which takes 50 texts of 5,000 characters a call\n- No backoff or Retry-After guidance for 429\n- The key can travel in a `Subscription-Key` query parameter\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 `Ocp-Apim-Subscription-Region` with the key for a regional resource\n- Pin `api-version=3.0` until your client handles the 2026-06-06 `inputs` and `value` shape\n- Batch up to 1,000 strings a call rather than one call per string\n- Spread load across the hour, since the character caps are hourly and bursts get 429\n- Treat 403001 as the free quota running out, not a permissions error\n\n## What the review panel asked for\n\n- Price the LLM option\n- Show every region's rate\n- when-not guidance\n- a v3.0 retirement date\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": "BB",
    "score": 72.2,
    "assessed": "2026-10-01",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 20,
        "maxGain": 10,
        "reason": "No x402, MPP or L402 (0). Per-million-character prices published without a login, readable through the Azure Retail Prices API since the pricing page needs JavaScript (20). F0 gives 2 million characters a month, but an Azure subscription needs a card (0). A person 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": "security",
        "name": "Security \u0026 auth",
        "score": 75,
        "maxGain": 4.4,
        "reason": "Model reading of the checklist, with training and retention in place of the least-privilege and injection lines. Resource keys, 10-minute bearer tokens, or Microsoft Entra ID with the Cognitive Services User role and managed identities (30), less 10 because the authentication reference documents passing the key in a `Subscription-Key` query parameter (20). The data privacy page says text translation doesn't store customer data, which leaves nothing to train on, but we didn't find an explicit training statement there (15 of 20). Text isn't stored and documents are hard-deleted after processing. LLM translation runs through your own Foundry deployment, whose retention terms we didn't check (12 of 15). Azure Monitor and the activity log record resource actions, per-request logging not confirmed (10 of 15). MSRC disclosure policy, Azure bounty programme, SOC 2 and ISO 27001 reports. The microsoft.com security.txt passed its Expires date on 2026-09-23 (18 of 20).",
        "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": "reliability",
        "name": "Reliability",
        "score": 83,
        "maxGain": 3.4,
        "reason": "Azure status page with post-incident reviews (20). No review in the last 90 days names Translator. One on 29 September 2026 covers intermittent failures for Azure OpenAI, Foundry and Cognitive Services in Sweden Central from 10:03 to 15:58 UTC, which may touch Translator resources and LLM deployments there, so we count it as minor. The public page lists only broad incidents (20). Limits published with numbers, 1,000 texts and 50,000 characters a request and hourly caps from 2 million characters on F0 to 200 million on S4 (15). 429 has three documented sub-codes and the limits page asks for load spread across the hour, but there's no Retry-After or backoff guidance for 429. Retry advice exists only for 408 and 503. Text translation has no side effects, so a retry is safe (8 of 15). The old Cognitive Services SLA address now redirects to Microsoft's Online Services SLA, a download we didn't open (10). API 2026-06-06 is GA (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": "ergonomics",
        "name": "Agent ergonomics",
        "score": 82,
        "maxGain": 2.9,
        "reason": "API reading of the checklist. Responses hold one translation per target with optional alignment and sentence lengths, and up to 1,000 texts go in one NMT call, but there's no field selection (20 of 25). Several targets per call, a `scope` filter on the language list, and `top` and `skip` paging on document jobs (18 of 20). Six-digit error codes such as 400036 (invalid target language) and 403001 (free quota exceeded) say what to fix (20). Text translation is stateless, but there's no 429 retry guidance and no idempotency key on document batches (12 of 20). A key, a region header and `api-version` on every call, with SDKs in Python, JavaScript, .NET and Java (12 of 15).",
        "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": "schema",
        "name": "Schema \u0026 documentation",
        "score": 83,
        "maxGain": 2.8,
        "reason": "OpenAPI documents for text translation 3.0 and 2026-06-06 are public in Azure/azure-rest-api-specs, with examples on each operation (25). No llms.txt found on learn.microsoft.com (0). Operation descriptions state purpose, and the overview says to pick NMT or an LLM by quality, cost and scenario, but not when to avoid either (15 of 20). Enums for `profanityAction`, `textType`, `tone` and `gender`, required fields marked, but the LLM `deploymentName` is a free string (13 of 15). Examples in the spec and a status-code page with six-digit error codes and their meanings (15). Dated `api-version` values and a What's New page (15).",
        "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": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 70,
        "maxGain": 2.6,
        "reason": "Read as a closed service. Document translation SDKs went GA in August 2026 (.NET 3.0.0 and JavaScript 1.0.0), after Python and Java 2.0.0 in July, and the text API 2026-06-06 in June (20, within 90 days). Four dated SDK releases in July and August (20). Public What's New page, Microsoft Q\u0026A and GitHub issue trackers on the SDK repositories, which we didn't read (10 of 15). Official SDKs current, `azure-ai-translation-text` 2.0.0 on 29 May 2026 for Python 3.9 to 3.13 (15). The SDKs live in the azure-sdk monorepos, CI not checked (5 of 10).",
        "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"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 80,
        "maxGain": 1.8,
        "reason": "Closed service under Microsoft's product terms, SDKs under MIT (15). The data privacy page and privacy statement agree on no storage of text and hard deletion of documents, but say nothing on training, and the LLM path falls under separate Foundry terms (22 of 30). Microsoft's lifecycle policy applies, but the What's New page gives no retirement date for v3.0 even though 2026-06-06 breaks its request shape (10 of 20). Regional resources and endpoints for data residency and a public Microsoft sub-processor list (18 of 20).",
        "blend": "editorial 65, provenance 95",
        "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"
      }
    ],
    "provenance": [
      {
        "label": "security.txt",
        "value": "published but past its Expires date",
        "points": 5,
        "max": 10
      }
    ],
    "unchecked": [
      "We didn't open the Online Services SLA document. The Cognitive Services SLA address redirects to it, which we took as coverage.",
      "unchecked: retention and abuse-monitoring terms for LLM translation through a Foundry deployment.",
      "The August 2026 SDK releases are dated by month only, so `lastRelease` stays at 2026-06-06."
    ],
    "weaknesses": [
      "The 2026-06-06 schema breaks v3.0 clients and drops BreakSentence and dictionary lookups from the spec",
      "LLM translation needs a Foundry deployment and bills tokens on the Azure OpenAI meter",
      "Tone and gender controls only on the LLM option, which takes 50 texts of 5,000 characters a call",
      "No backoff or Retry-After guidance for 429",
      "The key can travel in a `Subscription-Key` query parameter"
    ],
    "agentNotes": [
      "Send `Ocp-Apim-Subscription-Region` with the key for a regional resource",
      "Pin `api-version=3.0` until your client handles the 2026-06-06 `inputs` and `value` shape",
      "Batch up to 1,000 strings a call rather than one call per string",
      "Spread load across the hour, since the character caps are hourly and bursts get 429",
      "Treat 403001 as the free quota running out, not a permissions error"
    ],
    "requests": [
      {
        "text": "Price the LLM option",
        "reviews": 1
      },
      {
        "text": "Show every region's rate",
        "reviews": 1
      },
      {
        "text": "when-not guidance",
        "reviews": 1
      },
      {
        "text": "a v3.0 retirement date",
        "reviews": 1
      }
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-04",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.3",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}