Part of the Civic IQ MCP server

Vendor spend across 100,000+ agencies, one question away

See which vendors are already being paid in a category, and by which agencies, without exporting a spreadsheet first.

claude · connected to civic-iq-mcp

> Who's currently doing pavement management work in Ohio?

→ connecting to civic-iq-mcp...

Looks up vendors active in a category, filterable by state or agency

Same lookup works in reverse: start from a vendor name instead of a category

100,000+
agencies with spend data
Category or vendor
search either direction
Same data
as the Civic IQ platform your team uses
ClaudeChatGPTAPIAny MCP clientCivic IQMCP
How it connects

Spend data alongside meeting and contract signals, not in a silo

Spend history tells you who already has the relationship. Paired with board meeting monitoring from the same MCP server, you can see both the current vendor and the opening at once.

CL

Claude

Add the Civic IQ MCP server as a connector in Claude and ask about an agency, vendor, or category in a normal chat.

GPT

ChatGPT

Connect the same server as a custom connector or GPT action to bring agency signals into your existing ChatGPT workflows.

API

Custom agents

Point any MCP-compatible agent framework at the endpoint to pull structured signal data into your own tools.

What people ask it

Example queries teams run against this tool

Account teams use this tool to size a category and identify the current vendor before a call.

> Who are the active vendors for landscape maintenance in the Southeast?

Looks up vendors by category and region.

> Has this agency worked with a specific vendor before?

Checks for a vendor-agency relationship by name.

> What categories has this agency been active in recently?

Reverses the query: starts from the agency instead of the category.

> Search broadly for a vendor name if the category search comes back sparse

The search_vendor tool is built for exactly this fallback case.

What does the spend data cover?

Spend data covers 100,000+ agencies, a broader set than meeting coverage, because an agency's payments are published even when its meetings are not.

COVEREDCityCountyK-12Higher edDistrictsELSEWHEREFederal

SLED, not federal

This is state, local, and education spending: city, county, K-12, higher ed, and special districts.

find_vendorsCategory+ geographyVendorwho is paidsearch_vendor

Both directions

Query by category to find vendors, or by vendor to find everything held on it.

What can you ask it?

Plain language, the way you would brief a new rep. No query syntax, no tool names.

  • Which vendors are being paid for pavement management in Georgia?

    Vendors currently receiving payments in the category, by agency.

  • Who supplies biosolids hauling to agencies in New Jersey?

    Category to vendors across the state, so you know the incumbent before the call.

  • What has this county spent in my category over the last three years?

    Spend history for one agency, to size the account.

  • Show me everything you have on [vendor name] across all categories.

    The search_vendor fallback when a category search comes back sparse.

  • Who currently provides landscape maintenance to Gwinnett County, and are there any board meetings discussing rebidding it?

    Chains spend lookup with board-meeting signals: incumbent plus opening in one answer.

What comes back
SIGNAL RECORDAGENCYNorthside ISDMEETING DATE2026-04-14TOPIC MATCHutility submeteringSTAGEbudget approvedSOURCEboard minutes, item 7.2viasearch_signals

How is this different from open checkbook portals?

vs

Agency checkbook and transparency portals

Portals are one agency at a time, each with its own export format and its own idea of how to name a category. This is one query across the full SLED universe, with category and vendor names normalized so the same supplier is the same supplier everywhere.

vs

USAspending and federal databases

Those track federal awards. SLED spending, where school districts, cities, and counties actually buy, mostly never appears there, which is the gap this fills.

vs

FOIA and records requests

A request takes weeks per agency and comes back as unstructured PDFs. This is already collected, structured, and answerable in the chat window you have open.

Where else does this data show up?

FAQ

Government spending FAQ

See it running on your own accounts

Book time with our team and we'll walk through the Civic IQ MCP server live, using your target agencies as the example.

Book a walkthrough