Box API + MCP by Box

# Fix list: Box API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/box-api, the October 2026 research run, assessed 1 October 2026. Grade B, 69.6 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 Box API + MCP: 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, 25 out of 100, up to 9.4 more on the total

Why it scored 25: No x402, MPP or L402 (0). Plan prices public per seat; the price of API calls beyond the plan allowance is under Platform pricing, which we didn't read (10). The Individual plan is free with 10 GB, but the MCP server needs Business or above. We didn't confirm whether the free plan or the Business trial asks for a card (15). A person signs up in a browser and an admin enables the MCP server (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, 65 out of 100, up to 7 more on the total

Why it scored 65: Atlassian Statuspage at status.box.com with an RSS history (20). 20 entries between 7 July and 1 October 2026, among them upload errors for about two hours on 27 July, errors across multiple services for about two and a half hours on 6 August, and a search and metadata issue of about two hours on 7 July (5). 1,000 calls a minute a user, 240 uploads a minute a user, 6 searches a second a user and 12 an enterprise, per the 30 September check (15). 429 with a retry-after header, and SDKs that back off on their own, per the same check (15). We didn't find an uptime SLA this run (0). API and MCP server GA, the MCP server since August 2025 (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, 73 out of 100, up to 4.7 more on the total

Why it scored 73: OAuth 2.0 with scopes such as root_readonly and root_readwrite, short-lived access tokens and refresh tokens. The MCP server is an OAuth protected resource (28, two short because the MCP server asks for root_readwrite rather than letting a user pick read-only). Agents see only what their user can see, and 22 riskier tools are off until an admin enables them, but there's no documented confirmation step (15). The MCP server returns file text and Box AI answers over content other people shared, and the tools page has no prompt-injection guidance (3). Centralised audit logs and admin reporting (15). FedRAMP, HIPAA, PCI DSS and ISMAP on the security page. No bug bounty on that page and no security.txt on box.com (12).

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, 72 out of 100, up to 4.6 more on the total

Why it scored 72: The MCP server has 57 tools (5). An admin switches tool groups on and off and 22 start off, so the default surface is 35 (7 back). The API takes fields= to choose response fields, limit with offset or marker for paging, and filters on search (20). Documented JSON errors with a code, message and request ID (18). We couldn't check MCP annotations because the server code isn't published. The API takes If-Match and If-None-Match on updates, but has no idempotency keys (8). SDKs in Node, Python, Java, Windows (.NET) and iOS, and few required parameters (14).

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

Made of editorial 68, provenance 90.

Why it scored 79: Closed service with named contracting entities per region; the SDKs and OpenAPI spec are Apache-2.0 (18). Privacy policy with a revision effective 5 October 2026 per the 30 September check, and the security page says Box AI doesn't train on customer data (20). A written versioning policy, at least 12 months of support per version and 24 months between deprecation and retirement, with Deprecation headers (20). We didn't read a sub-processor list or data-location page this run (10).

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):

- security.txt: not found (0 of 10)

## 6. Schema & documentation, 91 out of 100, up to 1.5 more on the total

Why it scored 91: Public OpenAPI 3.0 spec in box/box-openapi with 297 operations, plus files for API versions 2025.0 and 2026.0 (25). developer.box.com/llms.txt indexes 400-odd pages, each with a Markdown twin (10). Operation descriptions state purpose and defaults, the shared link one for example explains each access level and what an empty object does, but rarely say when not to call (15). Typed parameters with enums such as open, company and collaborators, and required fields marked (13). Every operation documents error responses, parameters carry examples, but only 25 of 297 operations have response examples (13). Year-based API versions and a dated developer changelog (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, 84 out of 100, up to 1.4 more on the total

Why it scored 84: Changelog entries on 11 September 2026 (new Box AI models) and SDK releases on 9 September (30). Dozens of dated changelog entries since July (20). A public changelog and developer community for a closed service (12). Official SDKs in five languages released together on 9 September (15). The SDKs are generated from the public OpenAPI spec, which changed on 28 September. We didn't check SDK CI (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.

## 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 Box publishes an uptime SLA for paid plans
- unchecked: whether Box runs a public bug bounty; none is named on its security page
- unchecked: the price of API calls beyond the plan allowance (Platform pricing)
- The remote MCP server's code isn't published, so we couldn't read its tool descriptions or annotations
- The listing said about 54 tools and no llms.txt; the tools page now lists 57 and developer.box.com/llms.txt exists, so both were patched

## Weaknesses

- 20 status-feed entries between 7 July and 1 October 2026, two of them over two hours on uploads or multiple services
- MCP server needs Business or above, a three-seat minimum, and admin enablement per tool group
- The MCP server asks for root_readwrite, so a connected agent can write wherever its user can once tools are on
- API calls are metered per enterprise, 50,000 a month on Business
- No security.txt on box.com, and no bug bounty on its security page

## 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.

- Call who_am_i first; the tool list depends on the plan, the admin's toggles and the scopes granted
- Expect download and upload URL, move and shared-link tools to be missing unless an admin has enabled them
- Pass fields= to trim responses and page folder listings with limit and marker
- Send a box-version header to pin an API version, and watch responses for a Deprecation header
- For a link that expires, set shared_link.unshared_at on a paid account; the free plan can't

## What the review panel asked for

- Publish overage price
- Price AI units
- a read-only MCP grant
- confirmation on shared links

## 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.