Read AI

# Fix list: Read AI

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/read-ai, the October 2026 research run, assessed 8 October 2026. Grade D, 50.4 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 Read AI: 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, 52 out of 100, up to 9.6 more on the total

Why it scored 52: Graded as a hosted service on the MCP server and REST API. Statuspage at readai.statuspage.io with an API component and per-platform bot components (20). From 10 July to 8 October 2026 the API component has no incident. The bots have one critical incident (Zoom meetings not joined, 3 August, about three hours) and one major (23 July, under an hour), and Ask Read, which the API doesn't expose, has two majors (10). 100 requests a minute per user (15). 429 is documented as retry later with no Retry-After or backoff guidance, while create meeting agent and share meeting report are documented as safe to repeat (7). No SLA found (0). Both surfaces are open beta (0).

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, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Plan prices are public ($15, $22.50 and $29.75 a user a month billed annually) with nothing charged per call (10). The free plan needs no card and includes the API and MCP server (20). Client registration is programmatic, but a person has to sign in and consent 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, 54 out of 100, up to 8.1 more on the total

Why it scored 54: OAuth 2.1 with dynamic client registration, PKCE, 10-minute access tokens, rotating single-use refresh tokens and a revocation endpoint. Only two scopes, `meeting:read` and `mcp:execute`, so MCP access can't be narrowed to reading (24). A token reaches only reports its user can open, and a workspace admin can switch off access through the Downloads setting. No confirmation step is documented for sending a bot or sharing a report by email (9). Transcripts are untrusted speech and no injection guidance was found (0). Workspace audit logs in OCSF by webhook on Enterprise+ record sign-ins and integration connections, not individual calls (6). security.txt valid to 18 August 2027, SOC 2 Type 2 and a HIPAA option per the help centre, no bug bounty found (15).

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. Schema & documentation, 51 out of 100, up to 8 more on the total

Why it scored 51: No public OpenAPI file was found, and requests to api.read.ai/openapi.json and /docs got no response. The MCP tool schemas sit behind sign-in and were not read, so 5 of 25 for the published parameter tables. www.read.ai/llms.txt points to a help centre JSON API that returns every article (10). The API reference says when to use the live endpoint over the standard one, and the MCP article lists inputs, defaults and limits for each of four tools (13). Parameters are typed in tables and access levels are enumerated, with no schema to confirm constraints (8). Each endpoint has a curl request and a sample response. Errors are a table of status codes with no body schema (10). A /v1 path prefix, no changelog (5).

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.

## 5. Agent ergonomics, 63 out of 100, up to 6 more on the total

Why it scored 63: Four MCP tools, though list meetings returns the full detail of up to 10 meetings, transcripts included, per the MCP article. The REST API sizes responses with `expand[]` (20). Cursor pagination with `has_more`, a page size fixed at 10 or fewer and start-time filters only, with no search by title or participant (15). Errors are bare HTTP status codes with one-line meanings, though the 401 from the MCP endpoint carries a clear recovery message (8). Create meeting agent and share meeting report are documented as duplicate-safe. Tool annotations were not read (12). Few required parameters and sensible defaults, no official SDK (8).

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.

## 6. Maintenance & community, 31 out of 100, up to 6 more on the total

Why it scored 31: No changelog or release list was found, so dates come from the help centre. The MCP server article was last edited on 2 September 2026, 36 days before the check (20). Connection guides for Claude, ChatGPT and Microsoft Copilot Studio were published between 25 and 27 August 2026, which are documents and not dated releases (5 of 20). A support request form and a status page updated through October, no public changelog or developer forum found (6). Not in the official MCP registry on a search for read.ai, read-ai and readai, and no official SDK (0). No packages to assess (0).

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. Transparency & trust, 63 out of 100, up to 3.2 more on the total

Made of editorial 45, provenance 81.

Why it scored 63: Closed service with published terms, last updated 26 February 2026 (15). The privacy policy of 29 June 2026 keeps audio and video for no longer than two years and paid accounts' data until payment stops or deletion is requested. The help centre says model training is opt-in, while the terms grant a licence to train subject to opt-in or opt-out settings, and the DPA sits in a Trust Centre we couldn't read (20). No deprecation policy for the beta API (0). A sub-processor list is linked at trust.read.ai/subprocessors but didn't render for us, and EU and Canada data regions were announced on 2 February 2026 (10).

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: read.ai, registered 2017-12-16 (8 years) (11 of 15)
- Terms of service: read, states 6 of the 7 things a reader expects, and has 2 clauses that cost points (5.1 of 10)
- Changelog: not found (0 of 10)

## 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: the MCP tool definitions, input schemas and annotations, which need a signed-in Read AI account
- unchecked: the Trust Centre at trust.read.ai (SOC 2 report, DPA, sub-processor list), which renders only with JavaScript
- unchecked: whether api.read.ai publishes an OpenAPI file. Requests to /openapi.json and /docs got no response
- No changelog was found, so lastRelease is the edit date of the MCP server help article (2 September 2026) and may not match the last server change
- The webhook article gives two retry figures, up to 5 retries per request in the FAQ and 25 tries in the main text
- The date when create meeting agent and share meeting report were added to the MCP server was not established
- No date was found for general availability of the API and MCP server

## Weaknesses

- Both the REST API and the MCP server are labelled open beta, with no SLA and no deprecation policy found
- No static API keys or client credentials. A person signs in through a browser, and a broken refresh chain needs that person again
- No public OpenAPI file, changelog or official SDK was found. The reference is one help centre article
- List calls return at most 10 meetings a page, and the only filter is start time
- The bot can be sent to a meeting through MCP, but no tool removes it, reports its state or controls it during the call

## 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.

- Persist the new refresh token after every refresh. Access tokens expire after 10 minutes and each refresh token works once.
- Request only the fields needed with `expand[]` on the REST API. Transcripts and other expansions slow the response.
- Stay under 100 requests a minute per user and back off on 429. No Retry-After header is documented.
- Confirm with the user before calling create meeting agent or share meeting report. Sharing emails the recipient unless notify is set to no.
- Ask a workspace admin to enable Downloads in Reports & Sharing first. Without it the API and MCP server return nothing for workspace members.

## 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.