CloudConvert API

# Fix list: CloudConvert API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/cloudconvert, the October 2026 research run, assessed 9 October 2026. Grade B, 66.4 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 CloudConvert API: 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, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). The credit cost of each conversion type is published (1, 2 or 4 credits, plus 1 a minute), and the API page says prices start at $0.008 a file at 10,000 files, but the price per credit is drawn by a script and was not read, so 15 of 20. Free plan of 10 credits a day with no card (20). A person signs up in a browser to create a key or approve OAuth (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. Schema & documentation, 59 out of 100, up to 6.7 more on the total

Why it scored 59: No OpenAPI document was found in the docs or llms.txt. `GET /v2/operations` returns each operation's options with types, defaults and enum values, which is a machine-readable contract for options only. The MCP tool schemas sit behind OAuth and were not read (10 of 25). llms.txt and a Markdown twin of every docs page (10). The reference states what each endpoint does and when the synchronous variants are a poor choice, and the MCP tool list gives one line a tool (12). Task parameters are typed in the docs, but format options are found at run time through the operations endpoint and `createJob` takes an arbitrary task graph (10). Every operation has a raw request, CLI, cURL and SDK example, and the docs show 422, 429 and failed-task bodies with codes, with no full code list (11). The `/v2` prefix and pinnable engine versions, with no changelog found (6).

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.

## 3. Security & auth, 71 out of 100, up to 5.1 more on the total

Why it scored 71: API keys are created with any of six scopes and can be revoked, and OAuth 2.0 issues tokens for third-party clients, though the implicit grant is still supported and no rotation feature is documented. Signed URLs carry an HMAC signature, not the key (27). A key limited to `task.read` is read-only, and the MCP server asks for three scopes, one of them `task.write`. No confirmation step for deletes (12). Results come back as download links, and the `metadata` and `capture-website` operations handle untrusted content. The MCP page tells users to connect only the official URL, with no prompt-injection guidance (6). The security page says all access is logged and reviewable by the customer in activity logs, and jobs can be filtered by API key (11). ISO 27001 from TÜV Süd, a valid security.txt with a disclosure policy and no formal bounty, and no public advisories 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, 71 out of 100, up to 4.7 more on the total

Why it scored 71: The MCP server lists 24 tools, the 11 to 30 band, and the API returns links and status objects with an `include` parameter for the larger parts (16). List endpoints take `per_page` (100 by default, 1,000 at most), `page`, and filters by status, tag, operation and API key (17). Errors carry a `code` and a `message`, validation errors name the field, and failed tasks report codes such as `OPEN_FAILED` (15). No idempotency key. Jobs take a `tag`, a retry endpoint creates a new task, only successful conversions are charged, and the docs ask callers not to retry automatically. MCP annotations were not read (8). A job needs three short tasks, and there are official SDKs in six languages and a CLI (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. Reliability, 80 out of 100, up to 4 more on the total

Why it scored 80: Graded as a hosted service. Status page at status.cloudconvert.com with nine components and 90-day bars (20). Six incidents between 3 August and 8 October 2026, the longest readable ones a 23-minute slowdown in EU Central on 8 October and a six-minute multi-component alert on 7 September, with the API at 99.997 per cent over 90 days, so minor incidents only (20). Limits on creating jobs and tasks are described as dynamic, with `X-RateLimit-Limit` and `X-RateLimit-Remaining` headers and a free-plan cap of 5 concurrent tasks, but no fixed number per plan (8). A 429 carries `Retry-After` in seconds and the docs say CloudConvert retries retryable errors itself. No idempotency key (12). The terms promise 99.9 per cent monthly uptime with refunds on paid plans (10). API v2 is generally available and the MCP server carries 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.

## 6. Maintenance & community, 68 out of 100, up to 2.8 more on the total

Why it scored 68: The MCP server was announced on 3 September 2026, 36 days before this check (20). Five SDK and CLI releases since 11 July (CLI 3.0.1, Java 1.2.4 and 1.2.5, PHP 3.4.4, Ruby 1.2.0) and the MCP launch (20). No changelog. Support is a contact form and email, and dependency pull requests on the Java and CLI repositories were merged on 5 and 6 October 2026. Issue replies were not read (8 of 15). Six official SDKs, four released in 2026. The Node.js SDK's last tag is April 2025 and the Python SDK's is March 2022, and the MCP server is not in the official registry (12). Every SDK repository has a test workflow (8).

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, 75 out of 100, up to 2.2 more on the total

Made of editorial 64, provenance 85.

Why it scored 75: Closed service with dated terms that cover API use, and MIT SDKs (18). The privacy policy of 17 April 2026 gives 24 hours for files, 72 hours for account data after deletion and 180 days for logs, which agrees with the docs' 24-hour task deletion. It says files are not read, mined or copied and that AI is used only for the docs' question box. A DPA is signed on request (25). No deprecation policy. The terms give 30 days' notice only if the whole service is discontinued, and the operations endpoint flags deprecated engine versions (5). Hosting, payment, support and AI providers are named with addresses, and regions are Germany and Virginia, with no separate sub-processor list (16).

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 6 of the 8 things a reader expects (8.5 of 10)
- Changelog: 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 price per credit for packages and subscriptions. The pricing page draws it by script, and only the $0.008 a file starting figure on the API page was read
- unchecked: the MCP server's tool schemas and annotations, which need an OAuth sign-in to list
- unchecked: issue response times on the SDK repositories, and PyPI download counts
- unchecked: the ISO 27001 certificate itself. The security page links it and names TÜV Süd
- No OpenAPI document, changelog or deprecation policy was found in the docs, llms.txt or site footer
- The terms forbid offering a stand-alone raw file conversion service on the API, which a general-purpose agent operator should read before reselling conversions

## Weaknesses

- No OpenAPI document and no API changelog were found. The blog has one post since August 2024
- The price per credit is drawn by a script on the pricing page, so no price table is readable without a browser
- Rate limits are described as dynamic, with no fixed number per plan
- No idempotency key. The docs ask callers not to retry failed tasks automatically
- MCP tools take input files only as HTTP or HTTPS URLs, so local attachments fail in clients that do not turn uploads into links
- The terms forbid using the API to run a stand-alone raw file conversion service

## 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.

- Build each job from an import task, one or more processing tasks and an `export/url` task, then read the download links from the export task's `result.files`
- Call `GET /v2/operations` with `include=options` to find the options a format pair accepts before creating a `convert` task
- Do not retry a failed task automatically. CloudConvert retries retryable errors itself, and a retry endpoint creates a new task
- Download output within 24 hours. Tasks, files and export URLs are deleted after that
- Use `https://sandbox.api.cloudconvert.com/v2` for tests. It consumes no credits and accepts only a fixed set of approved test files

## 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.