Murf TTS API + MCP by Murf

# Fix list: Murf TTS API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/murf-tts, the October 2026 research run, assessed 1 October 2026. Grade BB, 70.9 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 Murf TTS 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). Falcon 2 at 1 cent per 1,000 characters and Gen2 at $0.03, published in the docs without a login (20). 100,000 free characters with no time limit behind a Get Free API Key button. We found no card step in the docs or FAQ (20). A person signs up in a browser to get the key (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, 68 out of 100, up to 5.6 more on the total

Why it scored 68: Model reading of the checklist, with training and retention in place of least-privilege and injection lines. One `api-key` header, plus a generate-token endpoint for short-lived client tokens. No key scopes found (23 of 30). The docs say customer text and audio never feed Murf's model training (20). Download links last 72 hours by the TTS overview and 24 hours by the Enterprise FAQ, and `encodedAsBase64` returns audio with nothing stored, so zero retention is self-serve but not the default (10 of 15). The dashboard shows daily and monthly usage, not per-call logs (5 of 15). SOC 2 Type II, ISO 27001 and ISO 42001 on the security page. No security.txt, disclosure policy 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. Agent ergonomics, 71 out of 100, up to 4.7 more on the total

Why it scored 71: Read as an API. Output in MP3, WAV, FLAC, A-law, mu-law or Base64 at 8 to 48 kHz, with word timestamps, streaming over HTTP or WebSocket, and inline Base64 instead of a download link (22 of 25). Per-request cap of 3,000 characters documented and the voice list filters by model, but there's no length control beyond splitting text yourself (12 of 20). Machine-readable error and warning codes with a `trace_id` (18 of 20). Backoff guidance documented (10). Nothing found on whether failed calls are billed and no seed for repeatable output (0 of 10). Three required fields including `locale`, and the only official SDK is Python (9 of 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.

## 4. Maintenance & community, 53 out of 100, up to 4.1 more on the total

Why it scored 53: Read as a model. Last changelog entry 2026-07-20 and `murf-mcp` 0.1.14 on 2026-08-20, inside 90 days (20). One changelog entry and one MCP release in the last 90 days (0 of 10). Gen2 streaming was deprecated on 2026-08-16 after 27 days' notice (0 of 10). Public changelog and a help centre (10 of 15). The Python SDK has no open issues (8 of 10). Official SDK in Python only, last released 2026-03-05 (8 of 15). Fern-generated with GitHub workflows, Python 3.8 and later (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.

## 5. Transparency & trust, 69 out of 100, up to 2.7 more on the total

Made of editorial 52, provenance 86.

Why it scored 69: Closed service under clear terms, SDK and MCP server MIT (15). Data statements disagree in places. The security page says all data sits in AWS us-east-2 with no storage elsewhere, while Falcon 2 runs in 12 regional environments and the subprocessor list names Azure, Google Cloud, IBM Cloud and Lambda Labs for speech synthesis. Download links expire after 72 or 24 hours depending on the page (12 of 30). Deprecations dated in the changelog, with no written policy and a short notice period (10 of 20). Subprocessor list with locations, last updated 2025-01-30 (15 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):

- Domain age: murf.ai, registered 2020-07-17 (6 years) (11 of 15)
- security.txt: not found (0 of 10)

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

Why it scored 85: OpenAPI and AsyncAPI specs listed in llms.txt (25). llms.txt with Markdown twins of every page (10). Model pages say which model suits voice agents and which suits studio renders, and the changelog flags Gen2 streaming as deprecated (15 of 20). Typed request bodies in the spec, with `text`, `voiceId` and `locale` required. We didn't check every enum (10 of 15). The errors page lists 7 HTTP codes, 4 WebSocket errors and 8 warnings, each with cause and fix, but covers only the streaming and WebSocket APIs (13 of 15). Dated public changelog, a single `/v1` path and one breaking rename in 2026 (`multiNativeLocale` to `locale` on 2026-03-05) (12 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.

## 7. Reliability, 90 out of 100, up to 2 more on the total

Why it scored 90: Atlassian status page at murf.statuspage.io, linked from the rate-limits doc, with history back to February 2024 (20). No incident posted since 2025-09-22, so the 90-day record is clean. The page tracks only two components (API and Studio), so a clean record says less than a page with a component per region (30). Concurrency limits published per model and region, 5 on US-East and 2 elsewhere for Falcon 2 on free and pay as you go (15). The errors page tells you to retry 429, 500 and 503 with exponential backoff (15). No SLA found on any self-serve tier (0). Falcon 2 carries no beta label (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.

## 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 a card is needed to claim the 100,000 free characters. The pricing page needs JavaScript and the docs don't say.
- Which download-link lifetime is right, 24 or 72 hours.
- Whether failed or truncated requests are billed.
- Whether the status page's API component covers all 12 Falcon 2 regional hosts.

## Weaknesses

- Falcon 2 concurrency is 2 outside US-East on free and pay as you go, 5 on US-East
- Gen2 streaming was deprecated on 2026-08-16 with 27 days' notice
- The security page says all data stays in AWS us-east-2, while Falcon 2 runs in 12 regions and the subprocessor list names four other clouds
- Download links last 72 hours per the TTS overview and 24 hours per the Enterprise FAQ
- Official SDK in Python only, last released 2026-03-05

## 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.

- Send `locale` on every request, since `multiNativeLocale` was replaced on 2026-03-05.
- Split text into chunks under 3,000 characters per request.
- Use the US-East host for 5 concurrent Falcon 2 calls instead of 2.
- Set `encodedAsBase64` to get audio inline and skip the expiring download link.
- Back off on 429 and close WebSocket contexts with `end: true` to free slots.

## What the review panel asked for

- Say whether the free key needs a card
- publish a failure-billing rule
- add a component per region

## 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.