SignalWire Voice API by SignalWire

# Fix list: SignalWire Voice API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/signalwire-voice, the October 2026 research run, assessed 1 October 2026. Grade B, 67.3 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 SignalWire Voice API: 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, 53 out of 100, up to 9.4 more on the total

Why it scored 53: status.signalwire.com redirects to a PagerDuty-hosted page that renders only with JavaScript, and we found no feed, so we couldn't read its components or history (10 of 20). No readable incident history (5). StatusGator, which can read it, shows outbound calls and fax degraded for about 2 hours on 14 August. Trial mode caps queued calls and messages at 10 each, and Space limits are raised on request. The error-codes page names a `rate_limit_exceeded` code but no general per-second figures (8 of 15). The error-codes page says to back off and retry on 429. The official Python SDK honours Retry-After with exponential backoff and retries POSTs only on 429 or 503, never on other 5xx, so a dial isn't replayed. No idempotency key (10 of 15). The Service Level Agreement (updated 1 January 2026) commits to 99.95 per cent monthly uptime for SignalWire Cloud APIs on every account, with a 10 per cent credit claimed by ticket within 30 days (10). The Calling API is generally available (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, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). US per-minute rates for calls, SIP, numbers, recording and the AI agent runtime published without a login (20). New accounts start in trial mode without a card, limited to 1 number and verified destinations. The trial page doesn't state a free credit amount (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. Agent ergonomics, 68 out of 100, up to 5.2 more on the total

Why it scored 68: A paging guide covers page sizes on list endpoints, with no field selection (15 of 25). Voice logs can be listed and paged. We didn't confirm the full filter set (15 of 20). An error-codes page with a structured `errors` array (type, code, message, attribute, url) an agent can map to fixes, and command results arrive on status webhooks (15 of 20). No idempotency key. The Python SDK's documented retry policy keeps POSTs from being replayed after a 5xx (8 of 20). One POST endpoint takes every command, a dial needs from, to and a SWML url, and there are official Python and Node SDKs plus Twilio-compatible ones (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. Security & auth, 72 out of 100, up to 4.9 more on the total

Why it scored 72: Basic auth with a Project ID and an API token, or short-lived Bearer tokens. Tokens are scoped when created, with the Voice scope for calls, the docs advise rotating them from the dashboard, and new tokens use a swapi_ format since September 2026 (30). Product scopes keep a calling token away from messaging, but we found no read-only token and no confirmation step for dialling (10 of 20). Calls carry untrusted caller speech, and the SWML ai verb passes it to a model. The platform webhooks page says SignalWire signs its requests with an HMAC key in `x-signalwire-signature`, and the Python SDK checks that signature on SWML fetches, SWAIG tool calls and summaries. The SDK's docs tell builders to expose only each step's tools, enforce rules in handlers and treat retrieved text as untrusted input, though there's no injection page in the platform docs (12 of 15). Voice logs can be listed by API. We didn't find an account audit log (10 of 15). A vulnerability disclosure policy, an ISO 27001 certificate and a SOC 2 Type 2 attestation are linked from signalwire.com/legal. 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.

## 5. Schema & documentation, 87 out of 100, up to 2.1 more on the total

Why it scored 87: Public OpenAPI 3.1 spec at signalwire.com/docs/openapi.json and .yaml, which generates the REST reference (25). llms.txt, and every docs page as Markdown by adding .md or sending `Accept: text/markdown` (10). The call-commands reference states what each command does, dial, calling.transfer, calling.record, calling.stream, calling.tap and calling.refer, with little on when not to use one (12 of 20). Typed fields from the spec, with commands carried in a params object (12 of 15). Examples per command and an error-codes page (13 of 15). Platform changelog with dated entries, such as the token format change on 2026-09-14 (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.

## 6. Transparency & trust, 78 out of 100, up to 1.9 more on the total

Made of editorial 65, provenance 90.

Why it scored 78: Closed service under the SignalWire Cloud Agreement (15). Privacy policy, a Data Processing Addendum and a cloud sub-processor list are published. We didn't find retention periods for call records or recordings (20 of 30). Dated deprecation notice for the token format, old tokens keep working (15 of 20). The sub-processor PDF names five cloud providers with countries (AWS in the US, Bahrain and South Africa, DigitalOcean in seven countries, Google in the US, Microsoft in four, Vultr in Australia), but service-specific sub-processors are listed without purposes or locations and the PDF carries no date (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):

- security.txt: not found (0 of 10)

## 7. Maintenance & community, 81 out of 100, up to 1.7 more on the total

Why it scored 81: signalwire-python 3.5.1 on 29 September 2026, and signalwire-js 4.0.0-rc.4 on 1 October. Changelog entry on 2026-09-14 for the new token format (30). Six Python SDK releases since 25 August (3.4.0, 3.4.1, 3.4.2, 3.4.3, 3.5.0, 3.5.1) plus two JS release candidates (20). Closed service with an active changelog and community support. We didn't sample issue replies (8 of 15). Official Python and Node SDKs current (15). The Python SDK runs test, multi-OS, nightly and docs-audit workflows and Renovate for dependencies. We couldn't see run results (8 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.

- unchecked: SignalWire's incident history for the last 90 days. The PagerDuty page and the feed path we tried both need JavaScript
- Rate-limit numbers beyond trial mode. The error-codes, paging and REST index pages name the 429 code without figures
- Retention periods for call records and recordings, which the DPA may state. We didn't open it
- The listing's lastRelease of 2026-09-30. The newest stable release we found is signalwire-python 3.5.1 on 29 September, with a JS release candidate on 1 October
- The listing had no OpenAPI URL. We've patched it to signalwire.com/docs/openapi.json and added the openapi tag
- The trial page says 1 number, not the listing's 2, and doesn't state the $5 free credit the listing gave. We've patched details, notable and pricingNotes

## Weaknesses

- Status page renders only with JavaScript, so its history isn't readable by agents
- No published rate-limit numbers beyond the trial's 10 queued calls
- No idempotency key, and the safe-retry rules live in the Python SDK rather than the API docs
- Per-customer Space hostnames, so there is no single public API host
- 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.

- Send dial with a url pointing at a SWML document, then use calling.* commands on the returned call ID
- Don't retry a dial on a 5xx. Retry only on 429 or 503, after Retry-After, or you may place the call twice
- Use your own Space subdomain as the host, there is no shared API host
- Create a token with only the Voice scope for a calling agent
- Verify `x-signalwire-signature` with the Signing Key before acting on a webhook

## What the review panel asked for

- State the trial credit amount
- serve the status page without JavaScript
- publish rate limits

## 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.