# Fix list: Sinch Voice API + MCP From Anchor Terminal's listing at https://www.anchorterminal.com/tools/sinch-voice, the October 2026 research run, assessed 1 October 2026. Grade B, 62.8 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 Sinch Voice API + MCP: 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, 52 out of 100, up to 9.6 more on the total Why it scored 52: Statuspage at status.sinch.com with calling and SIP components and history (20). One major in the last 90 days, delayed or failed in-app, PSTN and SIP trunk calling in AP-Southeast-1 for about 2.5 hours on 22 September, which IsDown marks major (10). No voice rate limits published. The v2 reference, re-read on 2 October, has no rate-limit section despite what the llms.txt summary suggests, and batch calls take a maxCps setting and a ttlSeconds window (0 of 15). The API docs don't cover 429, but the official SDKs retry it, honouring Retry-After and falling back to exponential backoff (Java 2.2.0 changelog), and v2 call, batch and service writes take an Idempotency-Key header that replays the cached response for 10 minutes (12 of 15). No SLA found (0). Voice API v1 is generally available. v2 carries no beta or preview label on its overview, reference or what's-new page, and the Node and Python SDKs ship it as standard (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. Security & auth, 50 out of 100, up to 8.8 more on the total Why it scored 50: Voice v2 takes a project access key over Basic auth or as an OAuth 2.0 token. v1 uses an application key and secret with signed requests recommended. We found no per-key scopes (25 of 30). No read-only mode or confirmation step found (0 of 20). Calls carry untrusted caller speech. Callbacks are signed in v1 and v2, and we found no prompt-injection guidance (5 of 15). Call lookup by API (the MCP's get-call-information uses it). No audit log found (10 of 15). The trust centre lists ISO 27001, SOC 2 Type II, PCI and HIPAA, annual penetration tests, and security@sinch.com for reports. No security.txt or bug bounty found (10 of 20). 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. Payments & pricing, 35 out of 100, up to 8.1 more on the total Why it scored 35: No x402, MPP or L402 (0). US outbound is in a downloadable price list, and the pricing page shows only illustrative inbound, SIP and number prices (15 of 20). Test credits and a test number on signup without a card (20). A person signs up in a browser (0). 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. ## 4. Schema & documentation, 79 out of 100, up to 3.4 more on the total Why it scored 79: The v2 reference links an OpenAPI file at developers.sinch.com/_bundle/docs/voice-2.0/api-reference/voice.yaml. Our reader got it as binary and couldn't parse it, so we read its shape through the SDK models generated from it (22 of 25). llms.txt with Markdown pages for v1 and v2 (10). Reference pages describe each SVAML action or v2 command, with a migration page on why to move to v2 (12 of 20). Typed fields with enums for call type, result and reason in the generated models (13 of 15). A v1 error and troubleshooting reference and request examples. We didn't find a v2 error page (12 of 15). Two API generations, v1 release notes last updated 2022-10-06 and an undated v2 what's-new page (10 of 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. ## 5. Agent ergonomics, 81 out of 100, up to 3.1 more on the total Why it scored 81: v2 lists take `page_size`, with no field selection (15 of 25). The v2 call list pages by link and filters by from, to, call type, time range, result and reason (18 of 20). v1 error reference with codes, and v2 errors carry a type and instance per the SDKs (15 of 20). v2 call, batch and service writes take an Idempotency-Key that returns the cached response for repeats within 10 minutes, and the webhook guide says to deduplicate on `ce-id` and `ce-source` (18 of 20). Official SDKs in Node, Python, Java and .NET, but a v2 dial is a nested commands document rather than a few fields (15). 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. ## 6. Transparency & trust, 73 out of 100, up to 2.4 more on the total Made of editorial 55, provenance 90. Why it scored 73: Closed service with general terms. The contracting Sinch entity is named only in the order form (15). Privacy notice, a DPA addendum by DocuSign, a sub-processor list and deletion on request. Sinch says it updates the sub-processor page at least 30 days before engaging a new one and emails subscribers. Recordings go to your own S3, Google Cloud Storage or Azure bucket. No retention periods found (22 of 30). No deprecation policy for v1 found (0). The sub-processor list, updated 28 September 2026, gives processing locations for most entries, a few without a region. For voice it names Microsoft Azure and Google Cloud for text to speech, AssemblyAI and Deepgram for speech to text, and TransUnion for STIR/SHAKEN (18 of 20). 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): - security.txt: not found (0 of 10) ## 7. Maintenance & community, 79 out of 100, up to 1.8 more on the total Why it scored 79: Node SDK 1.6.0, prepared on 23 September with Voice API v2 support, is the latest @sinch/sdk-core on npm, and Python 2.2.0 added the v2 voice domain on 21 September (30). Releases in the last 90 days include Java 2.1.1 (21 July), .NET 1.5.1 (2 September), Java 2.2.0 (17 September), Python 2.2.0 and Node 1.6.0 (20). v1 voice release notes stop in October 2022, and the SDK changelogs are dated (5 of 15). Official SDKs in four languages, current with v2 in Node and Python (15). CI on all four SDK repos, with CodeQL and dependency review on Java and an audit job on Node (9 of 10). The voice MCP's latest npm release is still 0.0.1-alpha.6 from 14 April, though its repo had commits on 29 September. 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. ## 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. - Voice API v2 carries no status label anywhere we looked. It's scored GA because nothing marks it beta and the SDKs ship it, not because Sinch says so - unchecked: the contents of the v2 OpenAPI file, which our reader still receives as binary. sinch/api-specs on GitHub stops in 2020 without voice, and the Java SDK repo carries generated models but no spec, so coverage is inferred from the models - No voice rate limits or SLA published. The 429 behaviour comes from SDK changelogs, not the API docs - The listing's lastRelease of 2026-09-23 matches Node SDK 1.6.0, the release that added Voice v2. It's kept ## Weaknesses - Two API generations with different hosts, auth and markup, and v1 release notes stop in 2022 - No published voice rate limits or SLA, and 429 handling is described only in SDK changelogs - Public pricing is a spreadsheet download plus illustrative examples - The voice MCP is an alpha (0.0.1-alpha.6) that can't stream audio - No security.txt or bug bounty ## 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. - Start new work on Voice API v2 at voice.api.sinch.com. v1 is the older callout model - Send an Idempotency-Key (a UUID v4) on v2 call, batch and service writes, and reuse it when you retry within 10 minutes - Use the MCP tts-callout tool for one-off spoken notifications, and the REST API for anything that streams audio - Trial accounts dial verified numbers only, one call at a time, so verify the test handset first - Verify the webhook signature and deduplicate on `ce-id` and `ce-source`, since fallback delivery can repeat events ## What the review panel asked for - Publish the price list as a page - publish voice rate limits - document 429 behaviour in the API docs ## 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.