Funraise MCP

Funraise MCP

Enables to interact with Funraise nonprofit fundraising platform, allowing read and write access to donor, donation, subscription, campaign site, household, and interaction data.

Category
Visit Server

README

Funraise MCP

MCP server for Funraise nonprofit fundraising - read and write your nonprofit's donor, donation, subscription, campaign site, household, and interaction data from Claude, Cursor, or any MCP client.

The first MCP for Funraise. The platform has 4.46 stars on Capterra with 114 reviews and powers thousands of mid-size to large nonprofits - the public REST API has had no first-party MCP wrapper and no Zapier/Composio/viaSocket integration.

You:   "How much did we bring in last month, and who were the top 5 new donors?"
Claude: *calls list_transactions with start_date, then list_supporters, summarizes the result*

You:   "Log this cash gift from the Patel family - $500, for the Year End campaign."
Claude: *calls create_supporter (if new) + create_transaction with the campaign site uuid*

You:   "Cancel Sarah's monthly subscription and email her a link to update her card instead."
Claude: *calls send_subscription_payment_update_email on the subscription, defers cancel*

Install

pip install -e .

Configure

# Get your API key in Funraise: Settings > API Keys > Generate API Key
export FUNRAISE_API_KEY="your-key-here"

The X-Api-Key header is the only authentication - no OAuth dance, no token rotation. One key per integration is the norm.

Plan limits

Plan Monthly requests Sustained rate Burst
Base (included) 3,000 1 req/sec 5 req/sec
API Plus 10,000 higher higher

The client retries on 429 with exponential backoff + full jitter, honoring the Retry-After header.

Use with Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "funraise-mcp": {
      "command": "funraise-mcp",
      "env": {
        "FUNRAISE_API_KEY": "your-key-here"
      }
    }
  }
}

Use with Claude Code

claude mcp add funraise-mcp -- funraise-mcp --env FUNRAISE_API_KEY=your-key-here

Tools (32 total)

Resource Tools
Diagnostic health_check
Supporters list_supporters, create_supporter, get_supporter, update_supporter, delete_supporter, print_yearly_summary
Transactions (one-time donations) list_transactions, get_transaction, create_transaction, update_transaction, move_transaction
Subscriptions (recurring gifts) list_subscriptions, get_subscription, create_subscription, cancel_subscription, send_subscription_payment_update_email
Campaign sites list_active_campaign_sites, list_archived_campaign_sites, get_campaign_site, publish_campaign_site
Forms list_forms
Households create_household, get_household, get_household_member_by_supporter, add_household_member, remove_household_member, merge_households, delete_household
Addresses get_address, create_address, delete_address, list_addresses_by_household, list_addresses_by_supporter, merge_addresses
Interactions (activity log) create_interaction, get_interaction, list_interactions_by_supporter, list_interactions_by_user, delete_interaction

Every tool raises on failure (so FastMCP sets isError: true on the wire response) and writes a JSONL audit record to stderr by default.

Use the Python client directly

import asyncio
from funraise_mcp import FunraiseClient

async def main():
    async with FunraiseClient() as c:
        # Find the most recent $5,000+ donation from this campaign
        txns = await c.list_transactions(
            limit=10,
            campaign_site_uuid="11111111-2222-3333-4444-555555555555",
            start_date="2026-06-01T00:00:00Z",
        )
        for txn in txns["items"]:
            if txn["amount"] >= 500000:  # cents
                print(f"Big gift: {txn['id']} - ${txn['amount']/100:.2f}")

asyncio.run(main())

Why this exists

  • First MCP for Funraise - the public REST API at https://api.funraise.io (v1 + v2) has had no first-party MCP server, no Zapier wrapper, no Composio toolkit, no viaSocket integration. The only LobeHub "Skill" is a Membrane CLI integration - not a real MCP.
  • Built for the conversations nonprofits actually have - "who gave last month?", "log this offline check", "send the Patels their 2025 giving summary", "merge these two duplicate household records". All one tool call away.
  • Production-grade engineering - shared httpx.AsyncClient with connection pooling + transport retries, typed exception hierarchy with structured fields, application-level retry with exponential backoff + full jitter honoring Retry-After, JSONL audit logging, py.typed marker, ruff + mypy --strict clean, 30+ respx-based tests, 100+ property-based tests via hypothesis.

API quirks worth knowing

  • Auth is one header - X-Api-Key: <key>. No OAuth, no token endpoint. Generate the key in Funraise > Settings > API Keys (you see it once - if lost, regenerate).
  • Versioned base URL - all endpoints live under /api/v2/. v1 is still served for some old endpoints.
  • Date fields are ISO 8601 - 2026-07-13T00:00:00Z. start_date/end_date filters on transaction lists accept these.
  • Resource ids - most use numeric ids (supporters, transactions, subscriptions, households, addresses, interactions). Campaign sites use UUIDs.
  • Errors - DRF-style {"detail": "..."} for 4xx, {"message": "..."} for some others, {"errors": {"field": [...]}} for validation. The client surfaces the most specific message on the typed exception.
  • One supporter, one primary address, one household - Funraise's "Household" is the family/address-level grouping. "Supporter" is the individual donor. A supporter can be a member of one household.

Development

git clone https://github.com/sanjibani/funraise-mcp
cd funraise-mcp
python -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"

# Lint + type-check
ruff check src tests
ruff format --check src tests
mypy src

# Run the test suite
pytest         # 50+ respx-based endpoint tests + property tests

# Run the MCP server
funraise-mcp

Architecture

Built on the same template (mcp-vertical-template) that ships hawksoft-mcp, open-dental-mcp, ezyvet-mcp, jobber-mcp, practicepanther-mcp, cox-automotive-mcp, qualia-mcp, realm-mp-mcp, campspot-mcp, cleancloud-mcp, playmetrics-mcp, foreup-mcp, storedge-mcp, courtreserve-mcp, procare-mcp, and passare-mcp. Industry-leading patterns from encode/httpx, stripe-python, boto3, pydantic, authlib, pytest-dev, astral-sh/ruff, and hynek/structlog.

  • Client (src/funraise_mcp/client.py) - async HTTP client with shared connection pool, transport retries, structured exception hierarchy, application-level retry with exponential backoff + full jitter.
  • Server (src/funraise_mcp/server.py) - FastMCP server with bare-raise error handling (so isError: true is set on every failure - the Blackwell audit pattern), JSONL audit logging on every tool call.
  • Exceptions (src/funraise_mcp/exceptions.py) - typed hierarchy (FunraiseAuthError, FunraiseNotFoundError, FunraiseRateLimitError, FunraiseAPIError, FunraiseConnectionError) with structured fields (http_status, error_code, request_id, retry_after).
  • Audit (src/funraise_mcp/audit.py) - stdlib-only JSONL audit logger with secret redaction, fail-open sink, env-var-overridable file path.

License

MIT.

See also

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured