Bandwidth Voice API + MCP by Bandwidth

# Fix list: Bandwidth Voice API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/bandwidth-voice, the October 2026 research run, assessed 1 October 2026. Grade B, 63.7 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 Bandwidth 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, 58 out of 100, up to 8.4 more on the total

Why it scored 58: Statuspage at status.bandwidth.com with datacentre and local-market components (20). IsDown counts 56 incidents in the last 90 days, 16 marked major. Most are single rate-centre impairments, which we count as minor, but two datacentre incidents hit voice for hours, LAX on 11 August (about 7 hours, outbound calls) and JFK on 14 August (about 5 hours, voice traffic). Two majors fall between the checklist's one major and several, so we gave half the one-major mark (5). Defaults published, 5 calls a second and 100 active sessions, with an outbound queue of about 5 minutes of CPS (1,500 calls at 5 CPS) (15). 429 is documented with two distinct messages for rate and concurrency, but no Retry-After or backoff guidance (8 of 15). No SLA found on the pages we read (0). Voice API is generally available. The MCP server is beta (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. 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 per-minute voice, recording, transcription and streaming prices are public, but number rental and SIP trunking are quoted by sales (15 of 20). Build trial with 3,000 credits, a US number 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.

## 3. Security & auth, 60 out of 100, up to 7 more on the total

Why it scored 60: OAuth 2.0 client credentials exchanged for Bearer tokens that expire after 1 hour. Each credential has roles that set its permissions and a basic or admin account scope, client secrets rotate with up to two active at once, and credentials can be deactivated or deleted (30). Roles and basic scope limit what an agent's credential can reach, though we found no read-only voice role and no confirmation step before dialling (10 of 20). Calls carry untrusted caller speech. We found no callback signing or prompt-injection guidance in the docs index (0 of 15). Voice Insights shows debug logs per call, including API request history and webhook delivery attempts, since 10 August 2026 (15). The legal page cites ISO 27001 and ISO 42001. No security.txt, bug bounty or disclosure policy found (5 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.

## 4. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: Call and recording lists take page sizes. No field selection (15 of 25). Filtering on call lists exists but we didn't confirm the full set this run (15 of 20). Errors page with codes and messages, and 429 bodies say which limit was hit (15 of 20). No idempotency key on call creation. Excess calls queue rather than fail, which makes a retry less often needed (5 of 20). Official SDKs in Node, Python and four more languages. A call needs to, from, applicationId and answerUrl, after an OAuth token exchange (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, 63 out of 100, up to 3.2 more on the total

Made of editorial 35, provenance 90.

Why it scored 63: Closed service with clear terms (communications services agreement). The MCP server repo has no licence file (15). Privacy notice, a data protection and privacy page, and a Data Protection Addendum inside the services agreement. No retention periods or sub-processor list found (15 of 30). The legacy credential migration is documented, with no dated deprecation policy found (5 of 20). No sub-processor list or data locations found (0).

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)

## 6. Schema & documentation, 88 out of 100, up to 2 more on the total

Why it scored 88: OpenAPI spec for the Voice API at dev.bandwidth.com/spec/voice.yml, with an OpenAPI downloads page (25). llms.txt that links Markdown versions of every docs page (10). BXML verb pages explain what each verb does and how callbacks follow, with less on when not to use one (14 of 20). Typed request bodies from the spec (13 of 15). A dedicated Programmable Voice errors page and request examples (14 of 15). Versioned /api/v2 path and dated release notes, though the release notes aren't linked from llms.txt (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. Maintenance & community, 78 out of 100, up to 1.9 more on the total

Why it scored 78: Release note on 2026-09-01 adding SIP trunking to the Build trial (30). Dated release notes on 2026-07-29, 2026-08-10, 2026-08-14 and 2026-09-01 among others (20). Closed service with active release notes and account management. The MCP repo has 2 stars and we didn't sample issue replies (8 of 15). Official SDKs in six languages, the Node SDK's latest being v8.1.4 on 2026-06-30 (15). SDKs are generated from the spec. CI not checked (5 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.

## 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 callbacks can be signed or only protected with Basic auth credentials
- The source of the listing's lastRelease of 2026-09-22, which we couldn't match to an SDK release
- Whether Bandwidth runs a disclosure programme or publishes a sub-processor list

## Weaknesses

- Number rental and SIP trunking prices aren't published
- No bug bounty, security.txt or callback signing found
- LAX and JFK datacentre incidents hit voice for about 7 and 5 hours in August
- No Retry-After or backoff guidance on 429
- MCP server is beta, installed by git clone, with no licence file

## 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.

- Get a token from the OAuth endpoint first, then POST to /accounts/{accountId}/calls with an applicationId and answerUrl
- Keep the call alive with Pause or a redirect while a bidirectional stream runs, or it hangs up when BXML runs out
- Read the 429 message to tell a rate limit from a concurrency limit before retrying
- Load the MCP server with BW_MCP_PROFILE=voice to keep the tool list short

## What the review panel asked for

- Publish number and SIP prices
- publish an availability SLA
- add Retry-After to 429

## 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.