# Fix list: TaxCloud From Anchor Terminal's listing at https://www.anchorterminal.com/tools/taxcloud, the October 2026 research run, assessed 8 October 2026. Grade C, 57.4 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 TaxCloud: 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. Security & auth, 45 out of 100, up to 9.6 more on the total Why it scored 45: API keys travel in the `X-API-KEY` header, never in the URL. An account can hold several, and the Management API lists, creates and deletes them. Keys made by API carry `audiences` with scopes, and account operations need scopes such as `create:accounts`. The docs give no list of scopes for ordinary keys, and a key from the web app covers both APIs by default (22 of 30). No read-only key was found. The bulk product delete needs a `confirm_delete` parameter, and test and live data are split by connection (7 of 20). Responses hold tax figures, addresses and the account's own records, with little third-party text (9 of 15). Responses carry `x-request-id`. No per-request log or audit log for the operator was found in the reviewed documentation (2 of 15). The security page describes Azure hosting, TLS, encryption at rest and daily backups. No security.txt, disclosure policy, bounty, SOC 2 or ISO 27001 statement was found (5 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. ## 2. Payments & pricing, 34 out of 100, up to 8.3 more on the total Why it scored 34: Read with the hosted rubric. No x402, MPP or L402 (0). All eight Premium tiers are public, from $79 a month for 200 orders to $1,299 for 40,000, with a stated rule that a calculation counts as a tenth of an order and overage is billed at the tier's price per order. That is plan pricing with a derivable unit price (14 of 20). A 30-day trial with Premium access and the API needs no card (20). A person has to register in a browser, complete a business profile and generate the key. Create an Account needs a service credential (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. Reliability, 66 out of 100, up to 6.8 more on the total Why it scored 66: Read with the hosted lines and scored on the v3 REST API at api.v3.taxcloud.com, the surface an agent would call. TaxCloud has no MCP server for the tax API. status.taxcloud.com redirects to a Datadog status page, created on 28 January 2026, with three components (20). The page is drawn by script. Its own config feed lists no incidents, and we could not read the uptime bars, so the 90-day record is scored a little short of clean (25 of 30). No rate limit numbers were found. The API page says "High threshold; subject to change based on usage" (0). Every operation lists 429, the docs advise exponential backoff on 500, TIC search documents 503 with Retry-After, refunds take an `idempotencyKey`, carts upsert and a duplicate `orderId` is refused. No backoff guidance for 429 was found in the docs, and order creation has no idempotency key (11 of 15). The security page says TaxCloud does not offer an SLA (0). v3 is the default version and is not marked beta (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. ## 4. Agent ergonomics, 64 out of 100, up to 5.9 more on the total Why it scored 64: Get Order takes `expand`, list calls and TIC search take `limit`, and one cart call can price several carts. There is no field selection (14 of 25). Certificates, access keys and products page by `cursor` with `limit`, `sortBy` and `ascending`, and products add `search` and `filters`. No list-orders operation exists, so orders are read one ID at a time (14 of 20). Errors are `application/problem+json` with `title`, `status`, `detail` and an `errors` list, confirmed on a 401, though the 401 body said only `Unauthorized` (16 of 20). Refunds take an `idempotencyKey` and return the original on replay, carts upsert by `cartId`, a repeated void returns 204 and account updates are safe to repeat. Create Order and cart conversion have no key, and the docs warn that converting a cart twice duplicates the liability (15 of 20). A cart needs `customerId`, `currency`, `origin`, `destination` and `lineItems`. No official v3 client library was found, and the API page points to OpenAPI Generator (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. ## 5. Transparency & trust, 51 out of 100, up to 4.3 more on the total Made of editorial 33, provenance 69. Why it scored 51: The service is closed under terms of service last updated March 2025, and the agent skills are Apache-2.0 (15 of 30). The privacy policy (1 February 2025) states no retention period. The terms let TaxCloud keep a copy of customer content where law, regulators or backups require it. No data processing agreement was found on the legal page (10 of 30). No deprecation policy was found, v1 is called legacy with no end date, and the terms allow policy changes with or without notice (3 of 20). The security page names Microsoft Azure, `Authorize.NET` and GoCardless. No sub-processor list or data location was found (5 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): - Terms of service: read, states 6 of the 7 things a reader expects, and has 4 clauses that cost points (1.1 of 10) - Privacy policy: read, states 5 of the 8 things a reader expects (7.8 of 10) - Changelog: not found (0 of 10) - security.txt: not found (0 of 10) ## 6. Schema & documentation, 75 out of 100, up to 4.1 more on the total Why it scored 75: OpenAPI 3.1 documents are public for the Sales Tax API, with 16 operations, and the Management API, with 14. The two links in the docs index return 404, and the working files are listed at docs.taxcloud.com/openapi.json (25). llms.txt and a Markdown copy of every page (10). Void, refund, TIC search and the account operations say when to use them, when not to and what a retry does. Six operations, among them four on exemption certificates and Get Order, have no description (13 of 20). Request bodies mark required fields, and the tax spec has 13 enums, 41 length limits and 14 numeric maxima. Line item TICs are bare integers (10 of 15). Reference pages carry request samples in several languages and a typed problem body for 400 to 500, and a guide gives causes and fixes for 400, 422 and 500. The spec files hold no examples (12 of 15). v1 and v3 are documented side by side with a migration guide, but both specs say version 1.0.0 and no changelog was found (5 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. Maintenance & community, 57 out of 100, up to 3.8 more on the total Why it scored 57: A closed service with no API changelog, so we read dated public artefacts. The agent skills repository was published on 24 September 2026, 14 days before the check, the WooCommerce plugin reached v8.4.19 on 6 October and the Magento module v1.5.0 on 1 October. None is a dated API change (22 of 30). The WooCommerce plugin alone has six tags since 15 August 2026 (20). Phone and email support come with both plans. No changelog or developer forum was found, and the WooCommerce repository shows 24 open issues, whose reply times we did not measure (6 of 15). No official v3 client library was found. The PHP library targets v1 and was last changed on 21 June 2024. The agent skills are current (3 of 15). The plugin repositories run verify and test workflows (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. ## 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 90-day uptime record on the status page. The page is drawn by script, and its config feed lists no incidents. - unchecked: whether 429 responses carry a Retry-After header, and what the rate limits are. We did not trigger a 429. - unchecked: which scopes an ordinary API key can carry. The API page says keys are assigned specific permissions, and the docs list none. - unchecked: whether the web app keeps a per-request API log. We did not sign up. - unchecked: whether the Canada coverage on the marketing site is reachable through the v3 API. The API docs describe US sales tax only. - The agent skills (24 September 2026) say keys from the web app lack Management API scopes and need support to provision one. The docs page on Management API access says web app keys include it by default. We followed the docs page. - The docs home page links a service level agreement article that redirects to an audit support article, and the security page says no SLA is given. - The lead was right on the vendor, docs and interface. It did not say that the OpenAPI links in the docs index return 404 or that API access needs the Premium plan. - The docs llms.txt and Markdown pages carry instructions addressed to AI agents. No deduction was taken and we did not act on them. - The terms forbid automated scraping of data in the service and use for benchmarking, which matters before our probes run. Recorded as a fact, with no deduction. - The API has no filing endpoint, so `tax.filing` was left out of the capabilities although TaxCloud sells filing in its web app. - No data processing agreement, sub-processor list, deprecation policy or changelog was found on the reviewed pages. ## Weaknesses - No rate limit numbers are published. The API page says only that the threshold is high and subject to change - The security page says TaxCloud does not offer an SLA, and the terms supply the service as is - No API changelog or deprecation policy was found, and the legacy v1 API has no published end date - No official client library for v3 was found. The vendor's PHP library targets v1 and was last changed in June 2024 - Each calculation or address check counts as a tenth of an order against the plan, test connections included ## 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-API-KEY` on every call and put the connection ID in the path, as in `/tax/connections/{connectionId}/carts`. A wrong ID returns 404 - Mark nexus states active in the account first. With none active the API returns zero tax and no error - Use a separate test connection. There is no sandbox, and orders on a live connection go into filing - Pass a stable `idempotencyKey` on refunds, and never convert the same cart twice. A second conversion creates a duplicate liability - Void a wrong order before 12:00 AM Eastern on the 10th of the month after its `completedDate`. Later requests return 422 ## 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.