Plancraft MCP Server
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Plancraft MCP Server list overdue invoices with customer and open amount"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Plancraft MCP server
Connect Plancraft to ChatGPT, Claude and any AI agent through Cotlet.dev. Work with contacts, projects, offers, invoices and working hours in your own signed-in account. Sending documents, recording payments and booking hours ask for approval by default.
What you can ask
List overdue invoices with the customer and open amount.
Draft an offer from my site notes in the selected project.
Prepare payment reminders and show me the list before sending.
Related MCP server: CoachingPortal MCP Server
Connect in 2 minutes
Get your API key at https://cotlet.dev/app and export it as COTLET_API_KEY.
ChatGPT
Settings → Apps & Connectors → Create, with connector URL https://cotlet.dev/mcp/v1/connect, then sign in with Cotlet.dev. No API key needed.
Claude
Settings → Connectors → Add custom connector, with URL https://cotlet.dev/mcp/v1/connect, then sign in with Cotlet.dev.
Claude Code
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"Cursor
Copy .cursor/mcp.json from this repo into your project:
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": {
"Authorization": "Bearer ${COTLET_API_KEY}"
}
}
}
}Run it locally (stdio)
This repo is a real MCP server. It speaks MCP over stdio, lists its Plancraft tools without any key, and forwards every tool call to the hosted Cotlet.dev server. The hosted URL above stays the recommended way. Use this when your client only starts local servers.
Run the included Python module with Python 3.10 or later:
python3 plancraft_agent.py serveOr with Docker:
docker build -t plancraft-mcp .
docker run -i --rm -e COTLET_API_KEY plancraft-mcpTools: plancraft_guide, plancraft_status, plancraft_open, plancraft_read, plancraft_click, plancraft_type, plancraft_close. Without COTLET_API_KEY a tool call answers: "Set COTLET_API_KEY. Get a key at https://cotlet.dev/app".
Command line
Python 3.10+, with no third-party runtime dependencies. Clone this repository and run:
export COTLET_API_KEY="your-key"
python3 plancraft_agent.py connect # ready config for ChatGPT, Claude, Claude Code and Cursor
python3 plancraft_agent.py status # is Plancraft ready in your Cotlet.dev account
python3 plancraft_agent.py guide # Plancraft screens and sign-in link
python3 plancraft_agent.py read # open Plancraft, read the page, close the browser
python3 plancraft_agent.py tools # list Cotlet.dev tools
python3 plancraft_agent.py serve # start the stdio MCP server (same as no argument)Questions from Plancraft users
Which account should I connect?
Open plancraft.com/de-de/login in your Cotlet.dev Cloud Browser and sign in with your own office user.
Which actions ask first?
Sending documents, recording payments and booking working hours ask for approval by default. You can change this per tool.
Is Cotlet affiliated with Plancraft?
No. Cotlet.dev is independent and not affiliated with Plancraft.
Learn more
Request access: https://cotlet.dev/signup?app=plancraft&utm_source=github&utm_medium=readme&utm_campaign=plancraft
Workflows on the hosted integration
The live integration page documents these office operations, selected from the live page:
list_contactscreate_contactlist_projectsget_projectcreate_projectlist_documentscreate_offercreate_invoice_from_documentsend_documentlist_invoicesrecord_paymentcreate_payment_reminderlist_time_entriescreate_time_entry
The local stdio adapter exposes seven browser helpers listed above. These helpers forward to Cotlet.dev; the workflow operations are documented on the hosted integration page.
Available Tools
7 toolsplancraft_clickB
Click a numbered element on the current Plancraft page.
Inputs: element (required).
plancraft_closeA
Close the Plancraft browser session when the work is done, so credits stop running.
Inputs: no named parameters.
plancraft_guideA
How to work in Plancraft (read projects, draft offers and invoices, review overdue payments and book working hours): its screens and the sign-in link. Call this first.
Inputs: task.
plancraft_openB
Open a Plancraft page in your signed-in Cloud Browser (default https://plancraft.com) to work with read projects, draft offers and invoices, review overdue payments and book working hours.
Inputs: url.
plancraft_readA
Read the current Plancraft page: its text and numbered buttons, links and fields.
Inputs: element.
plancraft_statusA
Check that your Cotlet.dev account and Plancraft are ready: account, credits and Plancraft connection.
Inputs: no named parameters.
plancraft_typeB
Type text into a numbered field on the current Plancraft page.
Inputs: text (required), element (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool maps to a distinct browser action or lifecycle step: open, read, click, type, guide, status, close. The only mild overlap is among the startup tools (open vs. guide vs. status), but their descriptions clarify that guide provides instructions, status checks readiness, and open launches the page.
All tools share a consistent plancraft_ prefix, and most names are concise action verbs (open, read, click, type, close). The lone noun form plancraft_status and the guide noun-ish form are minor deviations from the otherwise verb-based convention.
Seven tools is well-scoped for a browser-automation server covering page lifecycle plus interaction primitives. Each tool has a clear operational role, with no redundant or filler tools.
The core browser loop is covered: open, read, click, type, and close, along with a guide and connection-status check. Minor gaps exist for common browser actions such as scrolling, keyboard shortcuts, or waiting for dynamic content, but agents can likely work around them.
Maintenance
Related MCP Connectors
Create contacts, move deals, send proposals and invoices, read your inbox, from Claude or ChatGPT.
Connect ChatGPT and Claude to Fortnox: invoices, customers, reports. Every write needs your OK.
weclapp ERP in your AI assistant: reads instantly, writes only after a preview you approve.
AI agents for CRM, mail, calendar, ERP and accounting - you approve. EU-hosted, OAuth.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables ChatGPT, Claude, and other AI agents to read and manage clients, photography sessions, contracts, and invoices in a signed-in Iris Works account, including creating drafts and saving changes only after user approval.7Elastic 2.0
- AlicenseAqualityBmaintenanceEnables ChatGPT, Claude, and other AI agents to work with client check-ins, workout templates, and training programs in a user's signed-in account, with approval required before actions reach other people.7Elastic 2.0
- AlicenseAqualityBmaintenanceEnables ChatGPT, Claude, and other AI agents to draft estimates, schedule jobs, assign crew members, and review receivables in a user's signed-in business account, requiring approval before actions reach other people.7Elastic 2.0
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to work with estimates, quotes, job schedules, variations, and progress claims in a user's signed-in Buildxact account, with write actions requiring approval by default.Elastic 2.0