US healthcare data for AI agents: CMS, FDA adverse events, CDC, NPPES NPI. Keyless, real samples.
Scored across 8 tools
Most tools have clearly separable roles: list/search for discovery, sample_data for row-level evaluation, snowflake_trial_sql for mounting, and data_dictionary for table-level schema. The main overlap is get_product_details, which bundles price, docs, tables, and listing links—content that also lives in get_pricing and get_access_options—so an agent may hesitate between fetching one product's full record versus the focused pricing/access tools, though the descriptions do draw the lines.
Every name follows a clean snake_case verb_noun pattern: get_*, list_*, search_*. The verbs map intuitively to read-only operations and there are no style deviations or vague names.
Eight tools is well within the sweet spot and each covers a distinct step in the catalog workflow (discover, inspect, sample, price, access). No tool feels like filler or an arbitrary split of another.
The discover → evaluate → access lifecycle is well covered: listing, keyword search, product metadata, column dictionary, real sample rows, pricing, access paths, and trial SQL. Minor gaps remain, such as no explicit subscribe/purchase action or cross-product comparison, but agents can reach a decision and get started via the trial SQL.
Every way to get full access to a Dataplex product: Snowflake Marketplace, Databricks, free trial, docs, human contact.
Inputs: slug (required).
Column-level data dictionary for one Dataplex product: tables, columns with types and descriptions, temporal coverage, and validated SQL including the non-obvious query patterns (current-release filters, dedupe flags, text-encoded numbers). Pass table to get one table's columns; enough to plan an analysis before subscribing.
Inputs: slug (required), table.
Machine-readable pricing for all Dataplex products: monthly price, free-trial link, Snowflake Marketplace listing URL.
Inputs: no named parameters.
Full metadata for one Dataplex product: description, key facts, tables, temporal coverage, update frequency, feed or dataset catalog preview (CMS, CDC), price, docs, Snowflake Marketplace listing.
Inputs: slug (required).
100 REAL rows from a Dataplex product's primary table — the same governed views sold on Snowflake and Databricks. No signup needed. Use this to evaluate schema and content quality.
Inputs: slug (required), max_rows.
Copy-paste SQL for an agent INSIDE a Snowflake account to mount a Dataplex listing and start querying trial data in minutes — no browser, no sales call.
Inputs: slug (required).
List all Dataplex healthcare data products: CMS provider, facility, nursing home and prior authorization compliance data, FDA adverse events (FAERS/MAUDE), CDC public health statistics, NPPES NPI provider records, hospital and Medicaid spending. Returns slug, name, one-liner, price, and links for each.
Inputs: no named parameters.
Search Dataplex healthcare datasets by keyword (e.g. 'Medicare physician spending', 'nursing home compliance', 'prior authorization CMS-0057', 'FDA adverse events', 'NPI provider', 'CDC drug overdose deaths'). Matches product names and descriptions, 300+ CMS feed titles, 1,500+ CDC dataset names, and column names; returns matching products with the specific feeds, datasets and columns that matched.
Inputs: query (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Live US drug acquisition costs (CMS NADAC) for AI assistants. Free, no auth, weekly data.
US public data for AI agents: recalls, gov contracts, IRS ZIP income, TX leads, WHOIS, tech stack
HL7 FHIR R4 health-data API for AI agents: free sandbox, 30 synthetic patients, AI-ready context.
Pay-per-call US healthcare data: hospital financials, prices, quality, exclusions, wages.
Add one secure layer between your agents and this server.
