Basiq

# Fix list: Basiq

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/basiq, the October 2026 research run, assessed 9 October 2026. Grade C, 60.4 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 Basiq: 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. Payments & pricing, 20 out of 100, up to 10 more on the total

Why it scored 20: No x402, MPP or L402 (0). Unit prices are public, $0.50 per user per month for data, $0.25 for enrichment and from $3.00 a report, but a platform access fee has no published amount and plans run for at least 12 months. The pricing page is not linked from the home page and may be out of date, so half marks (10 of 20). The sandbox is free and self-serve with no card mentioned. There is no free live tier (10 of 20). A person registers in the dashboard and live access goes through sales (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.

## 2. Security & auth, 55 out of 100, up to 7.9 more on the total

Why it scored 55: API keys are named, many to an application, revocable in the dashboard and limited to chosen endpoints by permission sets, with a 90-day rotation guide and 60-minute bearer tokens (27 of 30). The `CLIENT_ACCESS` token travels in the query string of the Consent UI address, and the MCP page says headers can be passed as query parameters, so the checklist's 10 comes off (17 of 30). The service reads bank data and cannot move money, consent scopes are set per data type and MCP route groups can be switched off. Deleting a user or connection has no confirmation step, and web connectors mean Basiq stores bank logins, encrypted with AES-256 per connection (16 of 20). Responses carry bank-written transaction descriptions and no guidance on treating them as untrusted was found (7 of 15). `GET /events` and the webhook message log record activity. No per-request audit log was found in the docs (8 of 15). No security.txt, disclosure policy or bug bounty was found. A 2020 information security policy aims at ISO/IEC 27001:2013, the certifications named in the docs belong to the AWS data centres, and the accreditation logos on the data security page carry no text (7 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.

## 3. Agent ergonomics, 62 out of 100, up to 6.2 more on the total

Why it scored 62: Transactions take `limit` up to 500 and a `filter` on seven fields. No field selection or summary mode was found (15 of 25). List calls page with `links.next` and Insights lists hold at most 20 a page (17 of 20). Errors return a `correlationId`, a string `code`, `title`, `detail` and a `source` pointer, and failed job steps carry the bank's own message (17 of 20). No idempotency keys were found, so a repeated `POST /users` creates a second billable user. Reads are safe to repeat, and the five MCP tools set `readOnlyHint` and `destructiveHint` (8 of 20). Every call needs the `basiq-version` header and a token exchange with a non-standard Basic header, and no official SDK is published (5 of 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. Reliability, 70 out of 100, up to 6 more on the total

Why it scored 70: Graded on the REST API with the hosted lines. Statuspage at status.basiq.io with components for the APIs, dashboard, documentation, website and connectors, and a history feed whose 25 entries reach back to October 2024 (20). No incident was posted in the 90 days to 9 October 2026. The newest entries are from April 2026, when NAB began blocking the web connector, and that connector still shows a partial outage, which is the bank's doing (25 of 30). Numbers are published for some parts, 1,500 token requests per 5 minutes, 1,000 reports a day and 20 refreshes a day on an Open Banking connection, while the general limits are described without figures (10 of 15). A 429 with code `too-many-requests` is documented with the advice to retry after the limit resets. No `Retry-After` header, backoff timings or idempotency keys were found (5 of 15). No SLA was found, and the Terms of Service say Basiq does not warrant uninterrupted service (0). Version 3.0 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.

## 5. Maintenance & community, 48 out of 100, up to 4.6 more on the total

Why it scored 48: The newest changelog entry, Sep '26 Basiq Product Updates, read 9 days old on 9 October 2026, which puts it at about 30 September (30). Two changelog entries fall in the last 90 days, September and July, so the line for three is not met (0). A public changelog, support by email, chat and a service desk, and an issues link on the docs repository. Reply times were not sampled (9 of 15). No official SDK. The OpenAPI files are kept in a tagged repository (3 of 15). The docs repository runs lint, link-check and release workflows and merged dependency updates on 1 October 2026 (6 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.

## 6. Schema & documentation, 84 out of 100, up to 2.6 more on the total

Why it scored 84: Ten OpenAPI 3.0 files are public in the `basiqio-oss/Basiq-docs` repository and embedded in each reference page, 73 operations in all (25). llms.txt indexes the docs, reference and changelog, and each page has a Markdown twin (10). Every operation has a description except the nine webhook operations. Descriptions state the purpose and some say when a call applies, such as purge working only on active connections (14 of 20). Schemas use enums and required fields, with 167 enums in the Connect file, but `filter` is a free-form expression string and the token `scope` is an untyped string (10 of 15). Request and response examples are dense, and an error codes page lists codes by HTTP status (13 of 15). The version is pinned with the `basiq-version` header, the docs repository is tagged (v3.0.8 on 16 July 2026) and the changelog has ten entries for 2026, dated by month (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. Transparency & trust, 72 out of 100, up to 2.5 more on the total

Made of editorial 61, provenance 83.

Why it scored 72: A closed service with public Terms of Service that name Basiq Pty Ltd (15). The privacy policy, the CDR Policy and a complaints policy are public, and the CDR Policy says data is destroyed or de-identified within seconds of a consent ending. The privacy policy says collected content may be retained indefinitely after an account ends, which does not agree with the docs security page, and no DPA was found (17 of 30). The OpenAPI page says breaking changes come with major versions, a migration checklist covers version 3.0 and legacy endpoints are labelled. The ANZ New Zealand end-of-support notice carries no date and no notice period is stated (8 of 20). Storage in AWS Sydney and Melbourne, outsourced service providers in New Zealand and Serbia by name, and representatives allowed to store data in New Zealand and the United Kingdom are disclosed (16 of 20). Regulatory standing counts as an addition, as on other bank-data listings. The footer gives ACCC accreditation ADRBNK000208 (+5).

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 5 of the 7 things a reader expects, and has 1 clause that costs points (6.3 of 10)
- Privacy policy: read, states 4 of the 8 things a reader expects (7 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.

- The pricing page was reached by its address and is not linked from the home, product or about pages we read, so its figures may be out of date. It carries a 2026 footer.
- unchecked: the amount of the platform access fee and the currency of the listed prices. The Terms of Service say amounts are in Australian dollars
- unchecked: whether Basiq itself holds ISO 27001 or SOC 2. The accreditation logos on the data security page are images with no text, and Cuscal's site was not read
- unchecked: the changelog entries for May 2026 (CDR Insights API Updates) and March 2026, whose Markdown pages answered HTTP 500
- unchecked: exact dates of changelog entries. The page shows relative ages, so the last release date is derived from 9 days ago on 9 October 2026
- unchecked: the dashboard, which needs a login, so the request log, key permissions screen and MCP route switches are known from the docs alone
- unchecked: uptime percentages on the status page, which are drawn by script, and the status API, which robots.txt disallows
- unchecked: GitHub stars and issue reply times on the docs repository
- The MCP docs list seven tools and the server returned five to an unauthenticated client. Whether `fetch` and `search` appear with a credential or a paid docs plan was not established
- Which Consumer Data Right access model a new customer needs for live Open Banking data (representative, affiliate or own accreditation) was not read in detail
- One request to au-api.basiq.io for robots.txt returned the API's 403 JSON error, so the host has no robots file. No other request was sent to it
- Two addresses were tried without a link, www.basiq.io/pricing.html (200) and www.basiq.io/sitemap.xml (404)
- No clause on automated access, scraping or benchmarking was found in the Terms of Service

## Weaknesses

- Live data needs a plan with a 12-month minimum and a platform access fee whose amount is not published
- No SLA was found, and the Terms of Service say Basiq does not warrant uninterrupted service
- The privacy policy says collected content may be retained indefinitely after an account ends, while the docs security page says data is deleted at once
- No security.txt, disclosure policy or bug bounty was found, and the accreditation logos on the data security page carry no text
- No idempotency keys or `Retry-After` header are documented, and no official SDK is published. The docs say to generate one from the OpenAPI 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 the API key verbatim after `Basic` in the `Authorization` header of `POST /token`. Base64-encoding it returns a 400.
- Send `basiq-version: 3.0` on the token request, cache the bearer token for its 60 minutes, and stay under 1,500 token requests per 5 minutes.
- Send the account holder through the hosted Consent UI in a browser with a `CLIENT_ACCESS` token bound to their `userId`. The API alone cannot create a first connection on version 3.0.
- Poll `GET /jobs/{jobId}` until `verify-credentials`, `retrieve-accounts` and `retrieve-transactions` all read `success` before reading accounts or transactions.
- Follow `links.next` on `GET /users/{userId}/transactions`. A page holds at most 500 items, and pending transactions get new ids on each refresh.

## 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.