Avalara AvaTax

# Fix list: Avalara AvaTax

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/avalara, the October 2026 research run, assessed 8 October 2026. Grade B, 68.8 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 Avalara AvaTax: 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, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 in the developer docs, API reference or pricing page (0). Plan prices are public at $79 a state a month, or $69 with SST services, with larger accounts quoted and no price per transaction (10). A 90-day trial with no card, 1,000 transactions a day (20). Signup and the OAuth consent both need a person 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.

## 2. Reliability, 62 out of 100, up to 7.6 more on the total

Why it scored 62: Graded as a hosted service. Statuspage at status.avalara.com with 135 components, the AvaTax group split into Tax Calculation Service, Restv2, Batch, Reporting and others (20). Four incidents between 10 July and 8 October 2026. Three were short or outside AvaTax calculation. The fourth, on the Tax Calculation Service and AvaTax UI, ran from 22:32 PT on 5 October to 21:20 PT on 6 October, about 23 hours, marked minor with no cause given, so we scored between minor-only and one major outage (15). Numbers are published only for the trial (5 orders and 5 invoices a minute, 1,000 transactions a day) and report exports (5 a minute). Production limits not found (7). 429 is documented with three error codes, and the integration guide says to retry a timed-out invoice with CreateOrAdjustTransaction. No Retry-After or backoff guidance found (10). The Service Terms promise commercially reasonable efforts with no uptime percentage (0). REST v2 is GA and the MCP server is registered at 1.0.1 with no beta label (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, 65 out of 100, up to 6.1 more on the total

Why it scored 65: The MCP server uses OAuth with PKCE and a single scope, avalara_mcp. The REST guides lead with HTTP Basic using a username and password or an account ID and licence key. Keys are account-wide, can be named and are revoked on reset, and bearer tokens are accepted (20). Access follows user security roles, and a transaction is an unrecorded estimate unless an Invoice type is sent. No read-only mode or confirmation step is documented for the MCP's calculate_tax and create_nexus tools (10). Responses are mostly structured tax data with some merchant-entered text. No injection guidance found, and the Service Terms place responsibility for agent actions on the customer (8). An account audit endpoint for failed calls, audit log exports, and per-user attribution except with a licence key (12). Vulnerability disclosure programme, and SOC 1 Type 2, SOC 2 Type 2 and ISO 27001 listed for AvaTax. No security.txt and 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. Agent ergonomics, 84 out of 100, up to 2.6 more on the total

Why it scored 84: The MCP server documents ten tools. Their definitions sit behind OAuth, so size was not measured. On the API, `$include=SummaryOnly` or LinesOnly shrinks transaction responses (21). `$top`, `$skip`, `$filter` and `$orderBy` on 152 of 262 GET operations, with 1,000 records at most and `@nextLink` (20). Errors carry code, number, message, description, faultCode and helpLink (18). No idempotency key. CreateOrAdjustTransaction is the documented safe retry and CreateTransaction rejects a duplicate committed code. MCP annotations were not readable (10). CreateTransaction needs three fields (customerCode, date, lines) and defaults to an unrecorded SalesOrder estimate. SDKs in six languages (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. Schema & documentation, 85 out of 100, up to 2.4 more on the total

Why it scored 85: Public Swagger 2.0 file at sandbox-rest.avatax.com/swagger/v2/swagger.json with 491 operations and 635 models. MCP tool schemas sit behind OAuth and were not read (25). llms.txt plus a Markdown page per endpoint, 471 for AvaTax (10). 475 of 491 operations carry a description of more than 80 characters, several saying which call to use instead, such as CreateOrAdjustTransaction over CreateTransaction. MCP tool descriptions in the docs are one line each with a sample prompt (15). Enums on 203 properties, required fields on request models and examples on 2,717 properties, though `$filter` and `$include` are free-form strings (13). A curl example on each endpoint page and an error list of 429 codes, each with its own page (14). A written n.x.y versioning policy and a version number in the ping response, but the patch notes page stops at 23.4.1 on 26 April 2023 (8).

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.

## 6. Maintenance & community, 79 out of 100, up to 1.8 more on the total

Why it scored 79: Production answered as version 26.9.2 and the sandbox spec as 26.10.0 on 8 October 2026. The JavaScript and Python SDKs were tagged 26.9.0 on 16 September (30). SDK tags 26.7.3 on 27 July, 26.8.3 on 24 August and 26.9.0 on 16 September (20). Closed service. A developer community forum exists, but the public patch notes stop in April 2023, and we did not read reply times on the 24 open issues in the JavaScript SDK repository (7). com.avalara/avatax is in the official MCP registry under Avalara's own namespace, and the six SDKs are current (15). The JavaScript SDK has build and publish workflows and the Python SDK a publish workflow. Test results were not checked (7).

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, 79 out of 100, up to 1.8 more on the total

Made of editorial 69, provenance 88.

Why it scored 79: Closed service with versioned terms that keep every past version online (Service Terms 7.1, effective 1 October 2026, and AvaTax supplemental terms). SDKs are Apache-2.0 (17). The DPA (version 1.10, 2 June 2025) limits use of customer data to running the service, support and legal compliance, with retrieval for 10 days after termination and destruction on request up to 60 days after. No fixed retention periods, backups are excepted, and Avalara may use de-identified aggregate data for any lawful purpose (20). A written policy deprecates version n-2 when n ships and decommissions it 18 months later, and the spec dates field deprecations. The policy's example table has dates out of order, and the terms let documentation alone count as notice (15). Subprocessor list version 11.0 of 9 September 2026 with date added, purpose and country, hosting on AWS in the United States (17).

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 1 clause that costs points (8 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: the MCP server's tool definitions, input schemas and readOnlyHint or destructiveHint annotations, which need an OAuth sign-in with an Avalara account
- unchecked: production rate limits for transaction calls and whether 429 responses carry Retry-After. Neither was found in the public documentation
- unchecked: the trust centre's documents (security policy, AI governance) and the platform behind the vulnerability disclosure programme, which load only with JavaScript or on request
- unchecked: reply times on issues in the SDK repositories and whether SDK tests pass, since only the git history was read
- Whether an uptime SLA exists in order documents. The Service Terms give none, and the product page's 99.99 per cent figure is a marketing claim
- Whether the Claude Code tab on the MCP page gives a command. Only the Cursor configuration was readable without JavaScript
- The cause of the 5 to 6 October 2026 AvaTax degradation. The status page gives none and marks it minor
- The pricing page prices plans per state, so the unit prices are entered as monthly plans. No price per transaction was found

## Weaknesses

- Production rate limits are not published. The only numbers found are the trial limits and 5 calls a minute on report exports
- No price per transaction. Public plans are $79 a state a month, or $69 with SST services, and larger accounts are quoted
- The API patch notes page ends at version 23.4.1 of 26 April 2023, while production reports 26.9.2
- Tax calculation was degraded from 22:32 PT on 5 October to 21:20 PT on 6 October 2026, per status.avalara.com
- REST guides lead with HTTP Basic using a username and password or an account-wide licence key, and the MCP server has one scope, avalara_mcp

## 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.

- Omit `type` or send SalesOrder for an estimate. A type ending in Invoice is recorded, and `commit` set to true locks it for reporting
- Retry a timed-out invoice with CreateOrAdjustTransaction, not CreateTransaction, which errors when a committed transaction has the same code
- Use sandbox-rest.avatax.com with sandbox credentials for tests. Production is rest.avatax.com and the two accounts are separate
- Add `$include=SummaryOnly` or LinesOnly to shrink transaction responses, and page lists with `$top` (1,000 at most) and `$skip`
- Call /api/v2/utilities/ping first. It answers without credentials and shows whether the credentials sent were accepted

## 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.