You own the sales data pipeline at your company. It lives in Snowflake, and for months two groups have asked you for different things. Your reps want to ask questions in plain language and get contacts back. Your data engineers want to enrich records in bulk without leaving the warehouse. Today ZoomInfo announced that it has put ZoomInfo MCP and the ZoomInfo API inside Snowflake, and the announcement is specific about how the work splits between those two groups.
For your reps, there is a conversational layer. Snowflake CoWork and Cortex Agents call ZoomInfo's MCP server as an external connector. It comes with 14 tools in three groups. Finding covers Search Companies, Search Contacts, Find Recommended Contacts, Search Intent and Search Scoops. Enriching covers Enrich Companies, Contacts, Intent, Scoops and News. Researching covers Account Research and Contact Research.
For your data engineers, there is the API, called from Snowflake Tasks, Streams, Snowpark or external functions. The post quotes the bulk limit directly: "Bulk enrichment handles up to 10,000 records per request and 200,000 records per query." All of this sits alongside the Data Cubes ZoomInfo already ships to Snowflake Marketplace.
Before you plan anything, you sort the claims from the specifications. The post is careful here, and so should you be. The GTM Context Graph's "100M companies, 500M contacts, and billions of signals" is ZoomInfo's description of its own database. You can't check it from outside. The operational numbers are different. They are 25 requests per second on the Standard API, 33 on the Scaling API, OAuth 2.0, the user's existing role-based permissions carried into Snowflake, and audit logging on every call.
The permissions point is the one you keep. A rep in CoWork "only see[s] contact data their ZoomInfo seat already entitles them to," so the MCP connector doesn't quietly widen access. That is the question your security team will ask first, and the answer is no.
Next you look at what sits behind the 14 tools. The catalog shows how much larger the API estate is than the MCP. The ZoomInfo provider page lists 42 API pages. The Enrich, Lookup, Bulk and Monitoring endpoints in the story match the Company Enrich API, Contact Enrich API, Lookup API, Bulk API and Monitoring API/Webhooks API. The tiers it names include the Company Master Data Enrich API and the OrgChart Enrich API. The MCP's search and scoop tools sit over the Intent Search API and the Scoop Search API. The agentic-access profile counts 199 operations, of which 83 act on data and 5 are flagged human-in-the-loop.
Then comes the turn, and it is the part of the post that matters most to you. The catalog also holds a record of ZoomInfo itself, and that record isn't all flattering. ZoomInfo has a Kin Score of 87.2, in the exemplar band. Developer ergonomics at 92.9, access clarity at 92.1 and operational transparency at 92.1 carry it. Contract governance at 31.8 is its weak facet.
Agent Readiness is 64.3. The MCP server, rate-limit signal, delegated identity, protected resource metadata and dynamic client registration are all lit, and those match what the announcement describes. Two dimensions are unlit, and both matter for warehouse pipelines: idempotency and pagination.
That lands on your own design. You picture a Stream that re-fires a Task. The Task resubmits a 10,000-record enrichment job. The published contract doesn't say whether doing that twice is safe. You can't tell from the specification, and the post doesn't claim to know.
So here is what you take away. The announcement's operational facts are checkable and mostly strong: rate limits, OAuth 2.0, carried-over permissions, audit logging. Its database size is ZoomInfo's own claim. The catalog's record of the company agrees with the story where it can and flags the gap where it counts. Before you wire a Stream to a bulk enrichment Task, treat retries as an open question for ZoomInfo to answer, and design your jobs so a repeated submission can't do harm.
Read the original: https://apis.io/2026/11/13/zoominfo-brings-14-mcp-tools-and-its-api-into-snowflake/