Retell AI API + MCP by Retell AI

# Fix list: Retell AI API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/retell-ai, the October 2026 research run, assessed 1 October 2026. Grade B, 69.4 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 Retell AI 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, 60 out of 100, up to 8 more on the total

Why it scored 60: Statuspage at status.retellai.com with an incident feed back to 10 December 2025 (20). Five incidents since 3 July 2026, batch calls failing for 50 minutes on 14 July, inbound calls not connecting for 47 minutes on 29 July and 27 minutes on 7 August, web and phone calls disrupted for 69 minutes on 5 September, and 20 minutes of call disruption on 16 September. We count the 5 September incident as one major (10). 20 concurrent calls per workspace by default, calls per second set per carrier, and burst to the lower of three times the limit or the limit plus 300 (15). Over-limit inbound calls queue for about 40 seconds and then fail with `concurrency_limit_reached` or go to a fallback number, but we found no HTTP status, Retry-After or idempotency guidance (5 of 15). No SLA found for pay as you go (0). 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 of 40). Every component has a published per-minute price, infrastructure, voice, LLM and telephony (20). $10 of free credits with no card, per last week's check (20). Access starts with a human signup (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, 60 out of 100, up to 6.5 more on the total

Why it scored 60: The hosted MCP server exposes over 40 tools across 8 groups (5 of 25), and we add 5 because a key with read scopes limits what an agent can do through it (10 of 25). Call lists filter and page (16 of 20). Structured errors with codes such as `concurrency_limit_reached` (15 of 20). No idempotency keys and no tool annotations. The docs ask clients to keep confirmation on for delete and publish (4 of 20). Typed SDKs for Node.js and Python with few required fields to place a call (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, 76 out of 100, up to 4.2 more on the total

Why it scored 76: API keys can be created, deleted and rotated, and restricted to read or edit scopes for Build, Monitor and Deploy. Public keys for web calls can be limited to domains with reCAPTCHA (30). Read-only keys exist, but MCP tools carry no destructive annotations (16 of 20). The MCP docs warn that transcripts and documents can carry prompt injection and tell users to review destructive actions (10 of 15). Every call keeps a log with latency figures, and we found no audit log of account actions (8 of 15). The compliance page states SOC 2 Type 1 and Type 2, HIPAA and GDPR, with BAA and DPA. No security.txt or bug bounty found (12 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. Transparency & trust, 79 out of 100, up to 1.8 more on the total

Made of editorial 75, provenance 82.

Why it scored 79: Closed platform with clear terms and Apache-2.0 SDKs (18 of 30). Call data is kept indefinitely unless a per-agent retention period from 1 to 730 days is set, and the compliance page covers BAA, DPA and erasure. The statements agree, though keeping data forever is the default (22 of 30). Dated deprecation notices with an RSS feed, such as the legacy MCP server shutdown on 20 July 2026 (20). The 27 May 2026 changelog announced new subprocessors (Soniox, Ironclad). We didn't check data locations (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: retellai.com, registered 2023-06-17 (3 years) (7 of 15)
- security.txt: not found (0 of 10)

## 6. Schema & documentation, 92 out of 100, up to 1.3 more on the total

Why it scored 92: OpenAPI at docs.retellai.com/openapi-final.yaml (25). llms.txt with descriptive one-line summaries for each page (10). Pages say when to use a single-prompt agent and when a conversation flow (16 of 20). Typed request bodies with enums and required fields (13 of 15). Examples throughout and typed errors in the SDKs (13 of 15). A dated changelog, SDK releases and a deprecation-notice page with an RSS feed (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, 85 out of 100, up to 1.3 more on the total

Why it scored 85: TypeScript SDK v6.0.1 on 14 September 2026 (30). Eight SDK releases between 11 August and 14 September, plus a changelog post on 24 August (20). Public changelog and deprecation feed, support we didn't test (12 of 15 for a closed service). Node.js and Python SDKs current (15). SDK CI restored on 10 September (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.

- Which endpoints the TypeScript SDK 6.0.0 release removed on 14 September 2026, and whether the deprecation feed announced them.
- The exact tool count of the hosted MCP server, which the docs give as over 40.
- Retell's data locations, which we didn't check this run.

## Weaknesses

- Call data kept indefinitely unless retention is set per agent
- MCP tools carry no read-only or destructive annotations
- Web and phone calls disrupted for 69 minutes on 5 September 2026
- No SLA found below enterprise
- TypeScript SDK 6.0.0 removed endpoints on 14 September 2026

## 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.

- Give the agent a key with read scopes unless it has to change agents or place calls
- Keep MCP client confirmation on, the server marks nothing as destructive
- Set `reserved_inbound_concurrency` so outbound campaigns can't crowd out inbound calls
- Handle `concurrency_limit_reached` by retrying later or enabling burst at $0.10 a minute extra
- Set a data retention period on each agent that handles personal data

## What the review panel asked for

- add Retry-After to API limits
- publish an SLA for self-serve
- destructive hints on MCP tools

## 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.