ZoomInfo API + MCP

# Fix list: ZoomInfo API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/zoominfo, the October 2026 research run, assessed 9 October 2026. Grade BB, 72.2 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 ZoomInfo 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, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). Per-credit prices are public, $0.05 an AI credit and $0.38 a data credit at the $1,000 spend tier, with top-ups from $20 (20). A free account with 1,000 data credits and 1,000 AI credits needs no card (20). A person signs up and completes OAuth 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. Security & auth, 65 out of 100, up to 6.1 more on the total

Why it scored 65: OAuth 2.0 with PKCE or client credentials, 20 scopes, short-lived tokens, and client self-registration on the MCP server. No secret travels in a URL (28 of 30). Read and manage scopes are separate, admins switch API access on or off per user and cap credits per user, and Update GTM Context needs admin rights. Four MCP tools write, no confirmation step is built in, and there is no organisation-wide switch for MCP (13 of 20). News, scoops and conversation data are untrusted text, and no prompt-injection guidance was found (4 of 15). A usage endpoint, credit dashboards for admins and a `requestId` on errors. No per-call audit log found (8 of 15). ZoomInfo's `llms.txt` states SOC 2 Type II, ISO 27001, ISO 27701 and TRUSTe. The trust centre is drawn by script and was unread, security.txt answered HTTP 500, and no bug bounty was found (12 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.

## 3. Agent ergonomics, 71 out of 100, up to 4.7 more on the total

Why it scored 71: Graded on the REST API, with the MCP server noted. Enrich calls take an `outputFields` list and the CLI has `--select`, so responses can be sized. The MCP server lists 26 tools with no toolsets found (18 of 25). Search pages with `page[number]` and `page[size]` up to 100 records, with many filters and a cap at page 100 (18 of 20). Coded errors with a recommended action, a `retryable` flag and a table of recovery steps for agents (18 of 20). No idempotency keys found. Rejected 429 requests cost nothing and search is free, but a repeated enrichment is free only under Records Under Management, which the pricing page limits to Enterprise. MCP tool annotations could not be read without an account (10 of 20). No official SDKs, a JSON:API envelope on every request, and one official CLI (7 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.

## 4. Reliability, 87 out of 100, up to 2.6 more on the total

Why it scored 87: Graded with the hosted lines. Atlassian Statuspage at status.zoominfo.com with 16 components, among them ZoomInfo Enterprise API, Enrich and GTM Studio (20). The history page lists August to October 2026. It shows one major incident on GTM Studio audiences on 4 September lasting 93 minutes, an eight-hour audiences issue on 24 September marked no impact, a list-processing issue on 4 August, and none naming the API component (20 of 30). Limits published per package on three windows, from 5 a second on Builder to 35 on Scaling (15). 429 carries `Retry-After` and names the exhausted window, and rejected requests do not consume quota. No idempotency keys found for audience writes (12 of 15). The Licence Terms and Conditions guarantee 99.9 per cent monthly availability (10). API v1, MCP server and CLI carry 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.

## 5. Schema & documentation, 84 out of 100, up to 2.6 more on the total

Why it scored 84: Eight public OpenAPI 3.0 files, with 18 operations in the Data API file (25). `llms.txt` and Markdown twins on docs.gtm.ai and gtm.ai (10). Endpoint descriptions state when to search before enriching and how to find valid field names, and MCP tool pages mark superseded tools (16 of 20). The Data API file marks required fields 198 times but has 11 enums, because filter values come from Lookup calls (10 of 15). 150 examples in the Data API file and an error catalogue with codes and a recommended action each (13 of 15). The API is versioned as v1 with breaking changes promised only in a new version. The dated Release Hub covers the whole product with an API filter, and docs.gtm.ai has no changelog of its own (10 of 15). The credit page says rate-limit headers are not returned while the rate-limit page lists them.

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

Made of editorial 65, provenance 85.

Why it scored 75: Closed service with published terms naming ZoomInfo Technologies LLC, and an MIT CLI (15 of 30). The privacy policy covers people in the database and customers, names sources (web scanning, third parties, a research team, customer contributions and ZoomInfo Lite email access), a Data Protection Officer contact and a removal route. Retention has no fixed period, and the pricing page says both that credits never expire and that they expire after 12 months (22 of 30). Two rate-limit headers carry a removal date of 1 January 2027 and breaking changes are promised only in a new API version, but superseded MCP tools carry no dates and no deprecation policy was found (10 of 20). Sub-processor list updated 17 July 2026 with locations and roles (18 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):

- Terms of service: read, states 6 of the 7 things a reader expects, and has 2 clauses that cost points (5.1 of 10)
- security.txt: could not be fetched (0 of 10)

## 7. Maintenance & community, 76 out of 100, up to 2.1 more on the total

Why it scored 76: GTM CLI v2.1.1 on 2 October 2026, and docs pages edited as late as 8 October (30). CLI releases 1.1.0, 2.0.0, 2.1.0 and 2.1.1 since 26 August, plus Release Hub entries dated 15 September 2026 (20). A public Release Hub, a support address in the docs and a GitHub issue tracker for the CLI, whose handling we could not see (10 of 15). The CLI is current on npm and ZoomInfo's docs name listings in the Claude, ChatGPT and Perplexity connector directories. No official SDKs, and the MCP registry timed out so its listing is unchecked (8 of 15). The CLI repository has release workflows, semantic-release and npm provenance attestations (8 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.

## 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 trust centre at trust.zoominfo.com, which is drawn by script, so certifications rest on ZoomInfo's `llms.txt` and no bug bounty or disclosure policy was read
- unchecked: the official MCP registry, which timed out twice, so `registryName` is empty
- unchecked: MCP tool schemas and annotations, which need a signed-in session
- unchecked: security.txt, since www.zoominfo.com answered HTTP 500 for it
- unchecked: status history before August 2026, which the history page did not list
- Which terms a free GTM.ai signup accepts. The Licence Terms and Conditions govern ordering documents, and the gtm.ai footer links only the website terms
- Whether enrichment batches are 25 records (docs.gtm.ai) or 10 (gtm.ai MCP and CLI docs), and whether rate-limit headers are returned, since two pages disagree
- Whether credits expire. The pricing page says they never expire while the account is active and, in its FAQ, that they expire after 12 months
- The website terms of use forbid systematic or automated data collection without consent, while robots.txt allows the pages read. We read the legal pages, `llms.txt` and pricing as a visitor would, and nothing from the profile directory

## Weaknesses

- The Licence Terms and Conditions bar disclosing benchmark results and automated access outside ZoomInfo's own tools without written approval
- The documentation disagrees with itself. One page says rate-limit headers are not returned, another lists them. Enrichment batches are 25 on one site and 10 on the other
- No idempotency keys found, and ZoomInfo shows no cost preview before a paid MCP call
- No official SDKs. The docs direct developers to generate a client from the OpenAPI file
- Records Under Management, which makes repeat enrichment free for 12 months, does not apply to the free tier per the pricing 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 Lookup first and pass the returned IDs to Search. Unknown industry, metro or topic strings are rejected with a 422
- Search before Enrich. Search is free, and each newly enriched record costs one data credit
- Request only the scopes the task needs in the `scope` parameter of the token call. Omitting it grants every scope configured on the app
- On 429 read `X-RateLimit-Rejected-Bucket` and wait for `Retry-After`. An exhausted hourly window can mean a wait of several hundred seconds
- Turn off model training in the AI client before connecting the MCP server. ZoomInfo prohibits MCP data being used for training

## 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.