# Fix list: ElevenLabs Text to Speech API + MCP From Anchor Terminal's listing at https://www.anchorterminal.com/tools/elevenlabs-tts, the October 2026 research run, assessed 1 October 2026. Grade BB, 73.1 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 ElevenLabs Text to Speech 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. Payments & pricing, 40 out of 100, up to 7.5 more on the total Why it scored 40: No x402, MPP or L402 (0). Per-1,000-character prices for every model published without a login, from $0.04 on Flash and v4 Turbo to $0.08 on v4, v3 and Multilingual v2 (20). Free plan with 10,000 v4 characters a month and no 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. ## 2. Security & auth, 60 out of 100, up to 7 more on the total Why it scored 60: Model reading of the checklist, with training and retention in place of least-privilege and injection lines. Keys can be restricted to chosen endpoints and given a credit quota, the docs walk through rotation, and service accounts hold keys that don't belong to a person (30). The Terms (31 March 2026) let ElevenLabs use content to improve its models, with an opt-out under Data use in account settings that only applies going forward (10). TTS text and audio are kept by default and zero retention through `enable_logging=false` is Enterprise only (5). The generation history lists each request, though we found no audit log of key actions (10 of 15). A valid security.txt pointing to a disclosure address and report form. The trust centre at compliance.elevenlabs.io didn't render for us, and we found no bug bounty or public advisories (5). 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. Reliability, 80 out of 100, up to 4 more on the total Why it scored 80: incident.io status page with a Text-to-Speech component and dated history (20). Three incidents touched TTS in the last 90 days, all marked minor. Errors across US services for 9 minutes on 8 July, a provider error spike on 3 August that was fixed before it was posted, and about 6 hours of elevated TTS and STT latency on 26 August. The major incident on 29 September hit Agents, STT and the Creative Platform, not TTS (20). Concurrency per plan published with numbers, 4 on Starter up to 40 on Business (15). The error reference returns `rate_limit_exceeded` and `concurrent_limit_exceeded` as 429 and asks for exponential backoff (15). No SLA found for self-serve plans (0). The TTS endpoints are GA (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. ## 4. Agent ergonomics, 82 out of 100, up to 2.9 more on the total Why it scored 82: API reading of the checklist. MP3, PCM, WAV, Opus, mu-law and A-law output, streaming over HTTP or WebSocket, and character timestamps on a separate endpoint so you only pay the payload when you ask for it. Higher-quality formats are gated by plan (22 of 25). Per-request character caps published per model (40,000 on Flash v2.5, 10,000 on v4 and Multilingual v2, 5,000 on v3). We didn't confirm voice-list filters or paging in this run (10 of 20). Errors carry `type`, `code`, `message`, `param` and `request_id` (20). Backoff guidance for 429, and an optional `seed` for best-effort repeatable output. Nothing found on whether failed calls are billed (15 of 20). Only the text and a voice ID are required, `model_id` defaults to Multilingual v2 and `output_format` to 128 kbps MP3, and official SDKs ship in Python and TypeScript (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. ## 5. Transparency & trust, 71 out of 100, up to 2.5 more on the total Made of editorial 50, provenance 92. Why it scored 71: Closed service with clear terms and MIT SDKs (15). The Terms allow training with a forward-only opt-out, while the zero-retention page talks about third-party LLM providers not training, so a reader has to piece the TTS position together (15). Deprecated models are marked on the models page without dates or a written policy (5). EU, India and Singapore residency hosts documented, no named sub-processor list found (15). 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): - Domain age: elevenlabs.io, registered 2021-12-15 (4 years) (7 of 15) ## 6. Maintenance & community, 77 out of 100, up to 2 more on the total Why it scored 77: Read as a model. Newest changelog entry 2026-09-28 (30). Weekly dated entries, 16 in the last 90 days (10). Turbo v2 and v2.5 are marked deprecated in favour of Flash, but we found no dates or notice period (0 of 10). Public changelog and support channels (10 of 15). The Python SDK has 19 open issues, several opened in August 2026, and we couldn't see reply data (5 of 10). Official Python and TypeScript SDKs current (15). Fern-generated SDK with an open issue about a blank changelog (7 of 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. ## 7. Schema & documentation, 94 out of 100, up to 1 more on the total Why it scored 94: OpenAPI file at api.elevenlabs.io/openapi.json (25). llms.txt with a Markdown twin of every page (10). The models page says which model to pick for latency, quality or language count, though few pages say when not to use one (16 of 20). Request bodies typed, with enums for `output_format`, but `model_id` is a free string (13 of 15). Examples on each endpoint and an error reference with about 90 codes across 10 types, each with a cause (15). Versioned `/v1` paths and a dated changelog every week (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. ## 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. - Whether the voice list supports filters and paging an agent can use. - Whether failed or interrupted requests are billed. - Which certifications the trust centre lists, since it needs a browser. ## Weaknesses - Content may be used for training unless you opt out under Data use, and the opt-out only applies going forward - Zero retention (`enable_logging=false`) is Enterprise only, and TTS text and audio are kept by default - No SLA published for self-serve plans - Eleven v4 and v3 ignore SSML `` tags - v4 rises from $0.022 to $0.08 per 1,000 characters after 2026-10-12 ## 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. - Call Eleven v4 through Text to Dialogue, not `/v1/text-to-speech`. - Spell out numbers yourself on `eleven_flash_v2_5`, which doesn't normalise them by default, and turning that on is Enterprise only. - On 429 read `code`, back off on `rate_limit_exceeded` and wait for running requests on `concurrent_limit_exceeded`. - Keep each request under 10,000 characters on v4 and Multilingual v2, 5,000 on v3. - Give the agent a key scoped to Text to Speech with a credit quota. ## What the review panel asked for - State plan allowances after 12 October - state queue or reject behaviour - publish an availability SLA ## 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.