DataForSEO

# Fix list: DataForSEO

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/dataforseo, the October 2026 research run, assessed 8 October 2026. Grade B, 64.9 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 DataForSEO: 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: The REST API accepts one API login and generated password over HTTP Basic, covering every API on the account, with no scopes. The docs say credentials can't be sent in URL parameters. The hosted MCP server adds OAuth 2.0 with PKCE, dynamic client registration, refresh tokens and a revocation endpoint, with two coarse scopes (`api` and `profile`). We score between one all-powerful key and revocable tokens (18 of 30). The dashboard has an IP whitelist, daily cost limits by API and endpoint and duplicate-task limits. There is no read-only or per-API credential, and no approval step before a billed call (12 of 20). The API returns titles, snippets and page content from third-party sites, and no prompt-injection guidance was found in the docs or the MCP server's SKILL.md (0 of 15). Every response carries its cost, tasks can carry a `tag`, the ID List endpoint lists past tasks and the Errors endpoint lists failed tasks for 7 days (10 of 15). No security.txt (403 on the main site), and no SOC 2, ISO 27001 or bug bounty found in the home page, terms or privacy policy. The March 2026 website incident was reported in a public postmortem a day after the fix (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, 40 out of 100, up to 7.5 more on the total

Why it scored 40: Read with the hosted rubric. No x402, MPP or L402 found. Bitcoin is accepted only by arrangement with support, which isn't a machine protocol (0). Per-request prices are published without login for every API, such as $0.0006 a Google results page in the standard queue and $0.024 a Backlinks request plus $0.000036 a row (20). Registration adds a $1 credit that doesn't expire and the sandbox is free. No card requirement is stated for either (20). An account needs a person to register in a browser and verify an email address or phone number, and the OAuth flow ends with a person approving it (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, 84 out of 100, up to 3.2 more on the total

Why it scored 84: Read with the hosted lines and scored on the REST API v3. status.dataforseo.com is the vendor's own page with API V3, Dashboard, Website and 13 API components, and its feed returns daily incident records for up to 365 days (20). The feed shows no incident from 16 June to 8 October 2026. The last entry is a partial dashboard data problem on 15 June, and the last API incident a 25-minute partial outage of Live tasks on 23 April (30). Limits are published with numbers, 2,000 requests a minute in general, 30 simultaneous requests on database-backed APIs, 12 a minute on Live Google Ads, with `X-RateLimit-Limit` and `X-RateLimit-Remaining` headers (15). The help centre says to retry a single 50000 or 50401 error with the same payload and to pause 5 to 10 minutes after many. The rate limit arrives as code 40202 inside an HTTP 200, with no Retry-After, and there are no idempotency keys for task_post, which is billed (9 of 15). No SLA found. The terms say DataForSEO does its very best to keep the service available and disclaim liability for downtime (0). v3 is the only live version since v2 closed on 5 May 2026 (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, 84 out of 100, up to 2.6 more on the total

Why it scored 84: Appending `.ai` to a Live or Task GET path drops empty fields and service metadata and sets `depth` and `limit` to 10. The MCP server has four tools, uses `.ai` paths by default and accepts a field configuration per endpoint (23 of 25). Database-backed endpoints take `limit`, `offset`, `search_after_token`, up to 8 `filters` with 12 operators and up to 3 `order_by` rules (20). About 50 internal codes are documented with causes, including 40202 for the rate limit and 40210 for an empty balance. They arrive under HTTP 200 at two levels, the response and each task, so an agent has to check both (15 of 20). No idempotency keys. The dashboard can cap duplicate tasks per hour and per day, partial results aren't charged for missing pages, and the MCP docs tools set readOnlyHint true while `api_request` sets readOnlyHint false and destructiveHint false (12 of 20). A Backlinks call needs only `target`, a SERP call needs a keyword, a location and a language, and official clients exist for Python, TypeScript, C#, Java and PHP (14 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. Schema & documentation, 86 out of 100, up to 2.3 more on the total

Why it scored 86: An OpenAPI 3.0.1 file of about 4.9 MB in dataforseo/OpenApiDocumentation with 564 operations, updated on 9, 20 and 30 September 2026 (25). llms.txt at dataforseo.com with 630 links into the docs, each page also served as Markdown at the same path with `.md` (10). Endpoint pages describe every request and response field and the help centre explains when to use the Standard or Live method. llms.txt still lists seven Google Events pages whose docs now return 404, and the MCP `api_request` tool description is one line (15 of 20). Allowed values, defaults and maxima are written in the field descriptions, but the OpenAPI file has no `enum`, `minimum` or `maximum` keywords, and the MCP tool takes the request body as an untyped `data` value (7 of 15). Each endpoint page has a request example in cURL, PHP, Node.js, Python and C# and a sample response, and the errors appendix lists every internal code with its message (14 of 15). The version is in the path, and dataforseo.com/updates is a dated 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.

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

Made of editorial 65, provenance 83.

Why it scored 74: The API is closed under public terms of service updated 12 June 2026. The MCP server is Apache-2.0 and the Python client MIT (18 of 30). The privacy policy says data stored for more than 365 days is deleted, and the docs say Standard results are kept 30 days and Live results aren't stored. The two cover different data and don't conflict. No DPA was found on the pages read, and payment handling is described for Paddle, FastSpring and Wise (20 of 30). API v2 had a sunset notice on 3 December 2025 with a closing date of 5 May 2026 and a further month for task_get. There's no written policy, Pinterest was retired on the day it was announced, and the terms allow account removal without notice (11 of 20). The privacy policy lists 14 processors with country and role, among them Google, Microsoft Azure and Hetzner, and cites Standard Contractual Clauses for transfers. The location of customer data isn't stated (16 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):

- Domain age: dataforseo.com, registered 2016-10-24 (9 years) (11 of 15)
- Terms of service: read, states 4 of the 7 things a reader expects (7.4 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 82 out of 100, up to 1.6 more on the total

Why it scored 82: Two changelog entries on 6 October 2026, an organic result `checks` array and a Codex plugin (30). Ten dated entries between 15 September and 6 October, and MCP server releases on 11, 17 and 25 August and 2 October (20). Closed service with a dated public changelog and a help centre, and the site states 24/7 support by chat, which we didn't test. The MCP repository showed 4 open issues (12 of 15). Official Python and TypeScript clients reached 2.1.8 on 30 September 2026 with five releases since 11 August, alongside C#, Java and PHP clients. The vendor's MCP server isn't in the official MCP registry, where the only DataForSEO entry is a third party's wrapper (15). The client and MCP repositories have publish workflows only, with no test run in CI that we could see (5 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.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 26 to 27 March 2026. An attacker took over an admin session on the WordPress marketing website through the Flamingo plugin and deployed two webshells with command execution. DataForSEO says the site holds no customer data and no other infrastructure was affected, and published a postmortem on 28 March. Fixed and documented, so the deduction is reduced (-3). https://dataforseo.com/update/postmortem-wp-incident-march-27
- 16 September 2026. The Pinterest API and the Social Media API suite were retired with an announcement dated the same day, and the docs pages now return 404. We found no earlier public notice, though customers may have been told by email (-3). https://dataforseo.com/update/ending-support-pinterest-and-social-media-ap%d1%96

## 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 minimal commitment for Backlinks API access. The errors page mentions it under code 40204, and the help article it links to returned 404.
- unchecked: whether the API password can be regenerated or more than one credential issued, since the dashboard sits behind login.
- unchecked: whether registration asks for a card. The trial page doesn't mention one.
- unchecked: whether customers were told by email before the Pinterest and Social Media APIs were retired on 16 September 2026.
- unchecked: support response times. The 24/7 chat claim wasn't tested.
- No SLA, DPA, security certification or bug bounty was found on the home page, terms, privacy policy or docs. A trust page elsewhere on the site may exist.
- The status feed is the vendor's own, with no third-party monitor to compare against.
- https://dataforseo.com/.well-known/security.txt returned 403, recorded as none.

## Weaknesses

- The REST API takes one account-wide login and password over HTTP Basic, with no scopes or read-only credential
- No SLA found, and the terms disclaim liability for downtime
- Errors arrive as HTTP 200 with a `status_code` in the body, at both the response and the task level
- No security.txt, certification or bug bounty found, and the marketing website was compromised on 26 March 2026
- Pinterest and the Social Media API were retired on 16 September 2026, the day of the announcement

## 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.

- Check `status_code` on the response and on every task. HTTP 200 is returned for most errors, and 20000 or 20100 means success
- Append `.ai` to Live and Task GET paths to get the trimmed response. The MCP `api_request` tool does this unless `noAiMode` is true
- Build against https://sandbox.dataforseo.com/v3/ first. It is free and returns dummy data in the production structure
- Use the standard queue (`task_post`, then `task_get` or a `postback_url`) for bulk work. Live Google Ads endpoints allow 12 requests a minute
- Ask the account owner to set daily cost limits and an IP whitelist, since the credentials carry the whole prepaid balance

## 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.