Akoya

# Fix list: Akoya

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/akoya, the October 2026 research run, assessed 8 October 2026. Grade D, 48.3 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 Akoya: 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, 28 out of 100, up to 14.4 more on the total

Why it scored 28: Graded as a hosted service. No public status page was found. status.akoya.com did not connect, and the docs place network availability and provider outages inside the Data Recipient Hub, behind a login (0). No readable incident history (5). The only number is a recommended maximum of 5 calls a second on the Apps Management API, with none for the data APIs (5 of 15). The docs name 429 with code 1207 and say to slow bulk requests, and advise three retries with exponential backoff on 5xx. No Retry-After header and no idempotency key for app creation were found (8 of 15). No SLA is published. The home page's 99.9%+ network availability is a vendor claim, not a commitment (0). The v3 data APIs have been in production since 23 February 2026, with only the Tax product marked beta (10). Total 28.

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, 10 out of 100, up to 11.3 more on the total

Why it scored 10: No x402, MPP or L402 (0). The pricing page names Standard and Enterprise plans by monthly connections with no figures, and says a set-up fee may apply (0). The sandbox is free and self-service with test data and no card mentioned. There is no free live tier (10 of 20). A person requests a Data Recipient Hub account and sets up MFA, and production needs a security review and a signed agreement (0). Total 10.

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, 64 out of 100, up to 6.3 more on the total

Why it scored 64: OAuth 2.0 authorisation code grant with OpenID Connect. Tokens are issued per consumer and per app, ID tokens last 15 minutes to 24 hours, refresh tokens rotate, and `/revoke` ends a grant. Scopes are fixed, and access is narrowed by product subscriptions and the consumer's account selection (26 of 30). The data APIs are read-only, and the Hub has admin and viewer roles with mandatory MFA (17 of 20). The APIs return bank-written transaction descriptions, and no guidance on treating them as untrusted was found (7 of 15). Each response carries `x-akoya-interaction-id` for tracing. No per-call log for the operator was found (4 of 15). The security page states SOC 2 Type 2 and a security review of every participant. No security.txt, disclosure policy or bug bounty was found (10 of 20). Total 64.

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. Maintenance & community, 33 out of 100, up to 5.9 more on the total

Why it scored 33: The data API specification is marked "Updated 08/05/2026", read as 5 August 2026, and the newest changelog entry is about 31 July 2026, so the last dated change is within 90 days (20). Two dated changes in the last 90 days, short of three (0). A public changelog and a support centre with ticketing inside the Hub. No public community channel was found (8 of 15). No official SDK. A Postman workspace is the only client tooling (3 of 15). Six current OpenAPI specifications, with no packages to assess (2 of 10). Total 33.

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. Agent ergonomics, 67 out of 100, up to 5.4 more on the total

Why it scored 67: Responses can be sized with `limit` (default 50), `offset`, `startTime`, `endTime` and an `accountIds` filter. No field selection was found (17 of 25). Link-based pagination through `links.next.href`, though a small number of providers return no `prev` link (17 of 20). Errors carry a stable `code`, a `message` and an optional `debugMessage`, with causes documented. Code 602 covers both an expired token and missing consent (15 of 20). Every data endpoint is a GET, so calls are safe to repeat, and retry advice is documented. App creation has no idempotency key (14 of 20). v3 needs two or three custom headers and a provider ID on every call, the ID token stands in for an access token, and there is no SDK, only a Postman workspace (4 of 15). Total 67.

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. Schema & documentation, 70 out of 100, up to 4.9 more on the total

Why it scored 70: OpenAPI 3.1 specifications are published for all six APIs, as YAML code blocks to copy from docs pages with no raw file address (20 of 25). No `llms.txt` or Markdown docs. docs.akoya.com/llms.txt returns the HTML docs page (0). Each endpoint states its purpose, and guides say which product fits a use case, such as preferring customer information over `/contacts` (14 of 20). Parameters are typed and the data spec carries 73 enums, but `limit` is typed as a string and the docs warn that providers may return unknown enum values (10 of 15). The spec carries response examples, and an error reference lists codes with causes (13 of 15). The major version is in the path, with a version timeline and a changelog of eight entries since early 2025 (13 of 15). Total 70.

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. Transparency & trust, 58 out of 100, up to 3.7 more on the total

Made of editorial 44, provenance 72.

Why it scored 58: Closed service. The Terms of Use are public and dated 20 December 2024, while the data access agreement that governs production is not published (12 of 30). The privacy policy of 9 June 2025 gives no retention periods. The security page says Akoya passes data through without storing consumer financial data, while the docs say tokenised account number mappings are kept in a vault. No DPA was found (12 of 30). A version timeline gives release, deprecation and sunset dates with a six-month sunset rule. The Terms of Use still allow changes without prior notice (17 of 20). The privacy policy names Google Analytics. No subprocessor list or data location was found (3 of 20). Regulatory standing counts here as an addition to the checklist (+5 for a named regulator with a register number, +3 for a named regulator alone). Akoya names no regulator on the pages read (+0). Total 44.

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):

- Terms of service: read, states 7 of the 7 things a reader expects, and has 2 clauses that cost points (6 of 10)
- Privacy policy: read, states 2 of the 8 things a reader expects (5.5 of 10)
- Status page: not found (0 of 10)
- security.txt: 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: whether a status page exists at an address we did not try. status.akoya.com, trust.akoya.com and security.akoya.com did not connect from our network
- unchecked: the data access agreement that governs production use, which is sent through Docusign and not published
- unchecked: https://recipient.ddp.akoya.com/terms-of-use, named as the licence in the OpenAPI files, which returned a script shell
- The specification date "Updated 08/05/2026" is read in US order as 5 August 2026. Read as 8 May 2026, the newest dated change would be the changelog entry of about 31 July 2026
- The changelog shows relative dates only ("69 days ago"), so its entry dates are computed from 8 October 2026
- Whether the Standard plan has a list price. None is shown on the pricing page
- Whether section 7(b) of the Terms of Use, on automated systems, is meant to cover an AI agent calling the API under a data recipient's credentials
- The lead described Akoya as bank-owned. The pages read today do not state its ownership
- No sunset date is shown for v2 in the version timeline. The six-month rule would place it near 23 August 2026

## Weaknesses

- No price is published. Standard (under 10,000 monthly connections) and Enterprise are named without figures, and a set-up fee may apply
- Production access needs an onboarding questionnaire, a security review (SOC 2 report or questionnaire) and an agreement signed by Docusign
- No public status page, SLA or numeric rate limit for the data APIs was found. Availability figures sit inside the Data Recipient Hub
- The Terms of Use forbid using any automated system to access the network and allow changes and termination without prior notice
- No SDK, `llms.txt` or Markdown docs. The OpenAPI files are code blocks to copy from docs pages, not downloadable files

## 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 `x-akoya-interaction-type` (USER or BATCH) and `x-akoya-last-access` on every v3 data call, plus `x-akoya-intent-type` if the app subscribes to Payments
- Use the `id_token` as the bearer token. The `/token` response has no `access_token` field, so generic OAuth libraries need adjusting
- Treat ID tokens as valid for 15 minutes, refresh on error 602, and store the new refresh token returned by every refresh
- Page transactions by following `links.next.href` unchanged. Set `limit`, `startTime` and `endTime` on the first call only
- On 429 with code 1207 slow the request rate. On 5xx retry up to three times with exponential backoff

## 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.