# Fix list: Tink From Anchor Terminal's listing at https://www.anchorterminal.com/tools/tink, the October 2026 research run, assessed 8 October 2026. Grade B, 62.5 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 Tink: 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, 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 lists Standard as Contact us and Enterprise as Custom pricing, and the FAQ says there is no pay-per-use option (0). A free, self-serve Console account with sandbox test data and no card mentioned. No free live tier (10 of 20). A person signs up in the Console (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. Reliability, 62 out of 100, up to 7.6 more on the total Why it scored 62: Graded with the hosted lines. Statuspage at status.tink.com with 14 product components and incident history (20). In the 90 days to 8 October the feed lists one critical incident, on 16 July 2026, when authentication, consent, account data and payment flows failed in all European markets from 09:54 to 10:56 CEST, with a post-mortem published on 22 July. It also lists a 43-minute Payment Initiation incident on 25 September and 503 responses from Data Enrichment for 81 minutes on 6 October (10 of 30). Rate limits are stated as per app ID with no numbers (3 of 15). The reference documents the 429 status and an Idempotency-Key header with 24-hour keys, and tells customers to contact support on 429. No Retry-After or backoff guidance was found (9 of 15). Published SLAs of 99.7 per cent monthly uptime, a target under Basic Support and a commitment under service levels 1 to 3 (10). The v1 API and the Data v2 accounts and transactions endpoints carry no BETA tag (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. ## 3. Security & auth, 63 out of 100, up to 6.5 more on the total Why it scored 63: OAuth 2.0 client credentials sent in the POST body, scopes named per endpoint, 30-minute client tokens, two-hour user tokens and mutual TLS as an alternative. Secret rotation was not confirmed (27 of 30). Read scopes are separate from write and payment scopes, the account holder consents per bank in Tink Link, and one-time access data is deleted after 24 hours (15 of 20). Responses include bank-written transaction descriptions, and no guidance on treating them as untrusted was found (7 of 15). Every response returns an X-Request-ID and the Console has analytics. No operator audit log was found in the docs (4 of 15). The Privacy and Security Documentation describes controls aligned with ISO/IEC 27002, annual independent audits and penetration tests, and security.txt points to Visa's disclosure policy. No named certificate, bug bounty or security contact was found on the pages read (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. ## 4. Schema & documentation, 73 out of 100, up to 4.4 more on the total Why it scored 73: The API reference holds Swagger definitions for 205 operations, but only inside the docs app's JavaScript bundle. No downloadable spec was found, and Postman collections are published per product (10 of 25). llms.txt lists about 280 guides, each with a Markdown twin. The API reference is not in it (10). 191 of 205 operations have a description and the guides explain when to pick one-time or continuous access (14 of 20). Parameters are typed, 20 carry enums, and several older v1 endpoints take whole request bodies by reference (11 of 15). Guides carry curl requests and JSON responses, and each product has an error page with status and reason tables (13 of 15). The version sits in the path (v1, some v2) and the changelog has 367 dated entries (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. ## 5. Agent ergonomics, 74 out of 100, up to 4.2 more on the total Why it scored 74: Transactions return at most 100 a page and take date, account and status filters. No field selection was found (18 of 25). `pageSize` and `pageToken` on 20 operations, with snake-case equivalents on 6 more (18 of 20). Error pages list statuses and reasons with a `tracking_id`, and provider consent errors carry a `retryable` flag. API errors on Data v2 return a generic error object (16 of 20). Idempotency-Key is documented with 24-hour keys and appears on 7 write operations. Reads are safe to repeat (16 of 20). A first data call takes four token and user calls plus a browser consent, and the only SDKs found are Tink Link for iOS and Android (6 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. ## 6. Maintenance & community, 69 out of 100, up to 2.7 more on the total Why it scored 69: The newest changelog entry is 15 September 2026, 23 days before the check (30). Three dated entries in the last 90 days, on 4 August, 7 September and 15 September (20). Closed service with an active changelog and support through the Visa Support Hub. Basic Support carries no guaranteed response time, and GitHub issue replies were not sampled (8 of 15). Tink Link for iOS 5.1.0 shipped on 6 May 2026. The Android SDK's newest tag is 3.0.1 from 13 March 2025, with commits to 15 July 2026, and no server-side SDK was found (8 of 15). Neither SDK repository has a public CI workflow (3 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. ## 7. Transparency & trust, 90 out of 100, up to 0.9 more on the total Made of editorial 79, provenance 100. Why it scored 90: Closed service with a public Master Service Agreement dated 9 December 2025, governed by Swedish law. The mobile SDKs are MIT (15 of 30). A public DPA, a Privacy and Security Documentation and an end-user privacy notice agree on handling. Customer data is deleted within 30 days of termination and one-time access data after 24 hours. The end-user notice gives no fixed retention periods (24 of 30). The MSA commits to reasonable efforts at 90 days' notice before a product is discontinued, and the changelog carries dated deprecations, such as TLS 1.2 ending on 31 December 2027, announced on 31 March 2026 (17 of 20). The sub-processor list of 1 October 2025 names AWS and Google Cloud with EEA processing, and the MSA sets the data region as the EU, EEA and UK (18 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). Tink AB is authorised by the Swedish FSA under 44059 and Tink Financial Services Limited by the FCA under 988456 (+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. ## 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 changelog entry dated 7 September 2026 says redirect requests whose App URI doesn't match a registered Redirect URL are rejected from that same day. Whether customers had earlier notice was not established, so no deduction was made - unchecked: the Console itself (signup steps, whether a card is asked for, secret rotation, any audit log), which sits behind a login - unchecked: GitHub stars and issue responsiveness on the SDK repositories (the GitHub API refused us for rate limits) - unchecked: whether Tink holds a named certificate such as ISO 27001 or SOC 2. None was found on the pages read - unchecked: Visa's vulnerability disclosure page, which security.txt points to - The home page title says 6000 connections while the FAQ says 3,000+ in 18 countries and a docs page says 3,400+ - The US product has separate docs at docs.us.tink.com, which were not read. This listing covers the European platform ## Weaknesses - No public prices. The pricing page lists Standard as Contact us and Enterprise as Custom pricing, and the FAQ says there is no pay-per-use option - The API reference is a JavaScript app with the Swagger definitions inside its bundle. No downloadable spec was found - Rate limits are per app with no published numbers, and a 429 carries no documented Retry-After - A critical incident on 16 July 2026 stopped authentication, consent, account data and payment flows in all European markets for 62 minutes - security.txt has a Policy line pointing to Visa's disclosure page and no Contact or Expires field ## 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. - Request a client token at https://api.tink.com/api/v1/oauth/token, create a user, request an authorisation grant for that user, then exchange the code for a user token. - Renew tokens on a timer. Client tokens last 30 minutes and user tokens two hours, and the client credentials flow returns no refresh token. - Send the account holder through Tink Link in a browser. Bank consent cannot be completed by API calls alone. - Read transactions from /data/v2/transactions with `pageSize` up to 100 and follow `nextPageToken`. - Send an Idempotency-Key header on consent and payment writes. Keys are kept for 24 hours. ## 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.