fieldwork-mcp
Enables pest control business owners using Fieldwork to ask plain-English questions about customers, invoices, product usage, schedules, and technicians, returning clear answers from the Fieldwork API. Read-only access with no ability to modify jobs, customers, or payments.
README
Fieldwork MCP
Ask your pest control business questions in plain English.
Works with Claude, Cursor, Codex, and other AI apps that support MCP.
“How many customers do I have?”
“Who owes me money?”
“How much Alpine did we use this week?”
“Who did the most jobs?”
Your AI asks Fieldwork. You get a clear answer.
Read-only — it can’t change jobs, charge cards, or edit customers.
Site: landing/index.html (Relay home) · landing/fieldwork.html (Ask Fieldwork) · Roadmap: ROADMAP.md
Who this is for
Pest and lawn-care owners who run Fieldwork and want answers without digging through reports.
Why can’t they just ask Claude or Cursor to set it up?
Often they can — if they’re already in Cursor or Claude Code with this project open, say:
Set up Fieldwork MCP for me. I’ll paste my API key into
.env.
The agent can create the venv, write MCP config, and wire the launcher.
You still need to get the API key yourself (it’s your Fieldwork password-equivalent).
Don’t paste the key into a random chat if you can drop it straight into .env.
When a copy-paste snippet still helps
- You’re on the website and don’t have an agent in the folder yet
- You’re on Claude Desktop, which doesn’t edit your disk the same way
- Someone else is setting up a machine without opening this repo
The endgame (see Roadmap) is a one-line remote URL + login, like LandingFolio — no Terminal, no JSON.
What you can ask
Customers
- How many customers do I have?
- Do I have a customer named ___?
Money
- Who owes me money?
- How much did we make this month?
- How is business doing?
Product
- How much product was used?
- What chemicals do we carry?
Schedule, routes & technicians
- What’s on the schedule this week?
- Who are my technicians?
- How are my routes doing?
- Who did the most jobs?
Bonus prompt: monday_morning_briefing — one bundled ops snapshot.
Quick start
1. Get your Fieldwork API key
Easiest path (in the Fieldwork app):
- Go to Settings → Users & Routes
- Open your user (or an API Integration user)
- Open the Integration tab and copy the API Key
Help: Fieldwork APIs
2. Install
git clone https://github.com/daisydaines/fieldwork-mcp.git
cd fieldwork-mcp
python3 -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -e .
cp .env.example .env
# put FIELDWORK_API_KEY=... in .env
3. Connect your AI app
Cursor — ~/.cursor/mcp.json:
{
"mcpServers": {
"fieldwork": {
"command": "/ABS/PATH/TO/fieldwork-mcp/scripts/run_mcp.sh"
}
}
}
Claude Desktop — same JSON in
~/Library/Application Support/Claude/claude_desktop_config.json
Claude Code:
claude mcp add fieldwork -- /ABS/PATH/TO/fieldwork-mcp/scripts/run_mcp.sh
Codex — add a stdio MCP server pointing at scripts/run_mcp.sh (same idea as Cursor).
Then restart the app and ask: How many customers do I have?
Hosted connect (no Dropbox-style OAuth)
Fieldwork does not offer a third-party OAuth login popup. Closest path:
./scripts/run_mcp_http.sh
# open http://127.0.0.1:8000/connect
Paste your API key once. You get a bearer token and a one-line snippet for Claude / Cursor / Codex.
Also serves the Relay landing at /.
| URL | What |
|---|---|
http://127.0.0.1:8000/ |
Landing |
http://127.0.0.1:8000/connect |
Paste API key → connect token |
http://127.0.0.1:8000/mcp |
Streamable HTTP MCP (Authorization: Bearer <token>) |
For a shared deploy, set FIELDWORK_VAULT_SECRET (required) and FIELDWORK_PUBLIC_BASE_URL
(e.g. https://mcp.example.com). See Dockerfile. Keep FIELDWORK_MCP_ALLOW_ENV_FALLBACK=0.
Safety
| Can do | Cannot do |
|---|---|
| Read customers, jobs, invoices, usage | Create or edit customers |
| Summarize routes & techs | Cancel or reschedule jobs |
| Report product used on jobs | Take payments or change prices |
Tools are annotated readOnlyHint. The HTTP client refuses non-GET requests.
For builders
pip install -e ".[dev]"
pytest
ruff check src tests
./scripts/run_mcp.sh # stdio
./scripts/run_mcp_http.sh # streamable-http
Auth to Fieldwork: api-key header (default).
Unofficial. Not affiliated with Fieldwork / Anstar Products.
License
MIT — see LICENSE.
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.