DigitalOcean Spaces

# Fix list: DigitalOcean Spaces

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/digitalocean-spaces, the October 2026 research run, assessed 9 October 2026. Grade B, 63.2 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 DigitalOcean Spaces: 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, 20 out of 100, up to 10 more on the total

Why it scored 20: No x402, MPP or L402. Stablecoin payment through Stripe is an account payment method (0). Prices per GiB are public without a login (20). The $5 sign-up credit needs a valid payment method before any resource is created (0). A person signs up in a browser. The keys API needs a token from that account (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, 63 out of 100, up to 6 more on the total

Why it scored 63: The DigitalOcean API has a public OpenAPI file, which covers access keys and CDN endpoints. The S3 surface has no description file. Its reference lists each supported operation with headers and the minimum permission level, and defers to AWS for the rest (15). `llms.txt` and a Markdown twin of every page (10). The S3 reference gives one line per operation with a link to AWS, and the MCP tool descriptions are a few words each, such as 'Delete a CDN' (10). Header tables per operation, enumerated grants on the keys API, and MCP tools with required fields, a maximum and output schemas (10). A request and response example for most operations, but no error codes in the S3 reference beyond `NotImplemented` for unsupported calls (8). Dated Spaces release notes and a versioned DigitalOcean API. The limits page and the compatibility page disagree on `ListObjectsV2` (10).

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. Maintenance & community, 37 out of 100, up to 5.5 more on the total

Why it scored 37: Closed service. Its own release notes stop at 8 April 2026, 184 days before the check, which scores 0. The MCP server that carries the Spaces tools was tagged v1.2.0 on 7 October 2026 and its Spaces tools changed on 22 September, but that server cannot read or write objects, so it does not count as a release of the service graded here. No Spaces release notes in the last 90 days. The MCP server had seven tags in that time (10). Public release notes with an RSS feed, support tickets and a public repository with 193 commits in 90 days. Issue response times were not checked (10). Current official SDKs in three languages. The only MCP registry entry found for DigitalOcean is a third party's (10). The MCP repository has CI and release workflows and tests that guard tool annotations (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.

## 4. Reliability, 77 out of 100, up to 4.6 more on the total

Why it scored 77: Graded as a hosted API on the S3 surface. Status page at status.digitalocean.com with a readable history feed (20). The feed reaches back to 24 July 2026, 77 days. It shows failed write operations across services, Spaces among them, for 2 hours 26 minutes on 6 August, a 63-minute network disruption in BLR1 on 3 October that included Spaces, control panel access to Spaces failing for about two hours on 21 August and a Cold Storage billing display fault on 4 August. We read the 6 August event as one major incident (10). Limits published with numbers, 800 operations a second per new bucket and 5,000 requests an hour on the DigitalOcean API (15). Exponential backoff advised on `503 Slow Down`, and `retry-after` on the DigitalOcean API's 429. No retry or idempotency guidance for S3 writes (12). The Spaces SLA commits to 99.9 per cent monthly uptime per bucket (10). The S3 API and remote MCP servers are generally available (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.

## 5. Agent ergonomics, 73 out of 100, up to 4.4 more on the total

Why it scored 73: Graded as an API, with the MCP server noted. Responses are sized by S3's own controls, `Range` and `HEAD` among the listed headers, and the Spaces MCP endpoint carries only ten compact tools (20). S3 listing with prefix and marker, and `per_page` up to 200 on the DigitalOcean API. The limits page lists `list-objects-v2` pagination as a known issue (14). `503 Slow Down` and `NotImplemented` are the only S3 errors described. The DigitalOcean API documents its error bodies (10). A `PUT` by key can be repeated, conditional copy headers are listed, and all ten MCP tools set read-only, idempotent or destructive hints with a risk class (16). AWS SDK examples in several languages and official DigitalOcean SDKs for TypeScript, Go and Python. Most AWS SDKs need `us-east-1` as a placeholder region to create a bucket (13).

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. Security & auth, 77 out of 100, up to 4 more on the total

Why it scored 77: Access keys are revocable, can be regenerated and can be limited to named buckets with Read or Read/Write/Delete. DigitalOcean API tokens take custom scopes, and the MCP server issues short-lived OAuth tokens. Three off because SigV2 is still accepted and a limited key cannot coexist with a bucket policy (27). Read-only keys, versioning and a one-week recovery period for destroyed buckets. No Object Lock or confirmation step in the API was found (14). Returns stored bytes, with no guidance on treating them as untrusted. The key creation tool's description warns about handling the secret (8). Bucket and CDN access logs in S3 format, off by default, and an account security history that the docs say can show the wrong IP address for deletes (11). security.txt valid to 2028 with an Intigriti programme as contact, and SOC 2 and SOC 3 reports named on the trust page. ISO 27001 was not named on the page read (17).

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.

## 7. Transparency & trust, 74 out of 100, up to 2.3 more on the total

Made of editorial 63, provenance 85.

Why it scored 74: Closed service under a Terms of Service Agreement updated on 22 August 2026 that leaves ownership of content with the customer. The MCP server is MIT (15). The DPA of 6 February 2026 gives deletion within 30 days of account deactivation and breach notice without undue delay. The privacy policy excludes customer content, and the terms give no retention period in days (22). No general deprecation policy. The terms promise notice of changes as far as reasonably practicable, and old rate limits are dated per data centre (8). A sub-processor list of 10 June 2026 with activity and location, Cloudflare for the CDN and Wasabi for Cold Storage among them, and 13 named storage regions (18).

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):

- Endpoint on the vendor's domain: nyc3.digitaloceanspaces.com is not on digitalocean.com (0 of 15)

## 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 DigitalOcean OpenAPI file itself. Its address is listed in `llms.txt`, and we read the rendered Spaces Keys reference instead
- unchecked: the certification reports page and the Intigriti programme page, so whether a paid bounty exists and which SOC 2 type applies were not confirmed
- unchecked: status history before 24 July 2026. The feed holds 25 entries, which covers 77 of the 90 days
- unchecked: issue response times on the MCP repository, which a clone does not show, and npm download figures
- Legal, trust and SLA pages on www.digitalocean.com were read through a summarising reader, so clause wording should be confirmed before quoting
- Whether Cold Storage has a separate 99.5 per cent SLA, as the product details page says. The Spaces SLA page gives 99.9 per cent without naming tiers
- Whether `ListObjectsV2` pagination works. The limits page and the compatibility page disagree
- `lastRelease` is the MCP server tag of 7 October 2026. The Spaces release notes last changed on 8 April 2026
- The lead said Spaces is usable through remote MCP servers and the CLI. The MCP tools cover keys and CDN endpoints only, and the docs say `doctl` cannot manage files
- The Acceptable Use Policy bars scraping of service content and testing without written consent. Read as a fact with no deduction. It matters before any probe is run
- One typed address, https://www.digitalocean.com/legal/service-level-agreement, returned 404. The SLA was then found through the Service Terms

## Weaknesses

- The Spaces MCP server has ten tools for access keys and CDN endpoints. None reads, writes, lists or shares an object
- The access guide and the limits page say keys cannot be created through the API, while the API reference documents `POST /v2/spaces/keys`
- The limits page lists `list-objects-v2` pagination as unsupported, and the compatibility page says to use `ListObjectsV2` for new applications
- No error codes are documented in the S3 reference. Unsupported calls return `NotImplemented`, and no Object Lock was found
- The $5 sign-up credit needs a payment method first, and a subscription costs $5 a month once the first bucket exists
- The Acceptable Use Policy of 20 March 2026 bars scraping of service content and any probing or testing without written consent, which matters before any probe is run

## 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.

- Set the endpoint to `https://<region>.digitaloceanspaces.com` and, in AWS SDKs other than Python's, set `region` to `us-east-1` when creating a bucket
- Ask for a limited access key with a Read or Read/Write/Delete grant on one bucket. A full access key can create, configure and delete every bucket
- Do the object work through the S3 API. The MCP server and `doctl` cannot upload, list, move or delete files
- Retry with exponential backoff on `503 Slow Down`, and keep below 800 operations a second per bucket
- Fetch presigned links from the origin host, not the CDN. The docs say presigned requests through the CDN are not cached and can double the bandwidth charge
- Avoid many tiny objects. Each bills as at least 4 KiB on Standard and 128 KiB on Cold Storage, which also has a 30-day minimum

## 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.