Propublica Nonprofit MCP Server

ProPublica Nonprofit Explorer — IRS Form 990 Data

ProPublica's Nonprofit Explorer aggregates IRS Form 990 filings: every nonprofit's tax return data with revenue, expenses, executive compensation, grants given/received, and program services. ~1.8M nonprofits indexed since 2001. Free, no auth.

Part of Pipeworx — an MCP gateway connecting AI agents to 1856+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Why this matters for AI agents

For nonprofit research, "where does this charity's money actually go?" questions, executive compensation analysis, or tracing grant flows between foundations and recipients — Form 990 is the canonical data, and ProPublica makes it queryable. Pair with USAspending (federal grants), OpenFEC (political activity), and SEC EDGAR (corporate philanthropy).

Common flows:

• Org lookup. "Find Form 990 filings for X" → search by name or EIN.
• Org detail. Latest filing year financials: revenue, expenses, top compensation, program ratios.
• Grant tracing. "Who has the X Foundation funded?" → grants-given detail from Schedule I.
• Officer compensation. Top employees with salary, benefits, and other compensation.

Auth

None. ProPublica's Nonprofit Explorer API is fully public, free.

What Form 990 reveals

Common pitfalls

• 990 lag. Nonprofits file ~6-9 months after fiscal year-end. So 2024 fiscal-year data appears late 2025. Most-recent year is rarely current-year.
• Calendar vs fiscal year. Different orgs use different fiscal years. "FY 2023" may end in any month. Check the period-end date.
• 501(c)(3) vs (c)(4) vs (c)(6). All file 990s but with different rules and political-activity restrictions. The classification matters more than total revenue for understanding what an org does.
• Compensation reporting nuances. Top compensation includes salary + bonuses + retirement contributions + nontaxable benefits. The headline "compensation" number can vary depending on schedule used.
• Donor-advised funds (DAFs). Money flowing through DAFs (Fidelity Charitable, Schwab Charitable) appears as the DAF receiving and granting — the actual donor and ultimate beneficiary may be opaque.
• 501(c)(4) "social welfare." These can engage in some political activity but don't disclose donors. They show up here, but their political spending traces require OpenFEC cross-reference.
• Address normalization. Org addresses often have suite numbers, PO boxes, and registered-agent addresses that aren't where the org actually operates. Cross-reference geographic claims with state corporate registries.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "propublica-nonprofit": {
      "url": "https://gateway.pipeworx.io/propublica-nonprofit/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/propublica-nonprofit/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1856+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/search_nonprofits \
  -H 'Content-Type: application/json' \
  -d '{"query":"red cross"}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/search_nonprofits. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Run it from npm

@pipeworx/mcp-propublica-nonprofit is a local stdio bridge to the Pipeworx gateway: your MCP client starts it with npx, and it forwards every call to the same https://gateway.pipeworx.io/propublica-nonprofit/mcp endpoint shown above. It needs network access; there is no offline mode.

{
  "mcpServers": {
    "propublica-nonprofit": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-propublica-nonprofit"]
    }
  }
}

Without a key it calls anonymously, capped at 50 calls a day. A free account raises that to 200 a day: get a key at https://pipeworx.io/signup and add "env": { "PIPEWORX_API_KEY": "<your key>" } to the server entry. The key is sent as a Bearer token, the same way a hosted connection sends it, so calls are metered and billed exactly as they are there.

It lists only this pack's tools by default. Set PIPEWORX_META_TOOLS=1 to also list ask_pipeworx and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in fetch does not read HTTPS_PROXY by default. Newer Node releases honour it when NODE_USE_ENV_PROXY=1 is set; check your Node version's documentation.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Propublica Nonprofit data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides
• pipeworx.io

License

MIT