Invompt MCP
Click on "Install 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., "@Invompt MCPCreate a professional invoice for 8 hours of design work at $125 per hour"
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.
invompt-mcp
MCP server for Invompt's public invoice API.
It exposes:
invompt://spec/iml/v1as a cached MCP resourcecreate_invoiceas the write tool for invoice creationdraft_invoice_imlas an optional prompt for generating raw IML YAML
Quickstart
NPX
{
"mcpServers": {
"invompt": {
"command": "npx",
"args": ["-y", "invompt-mcp"],
"env": {
"INVOMPT_API_KEY": "inv_sk_your_key_here"
}
}
}
}Bunx
{
"mcpServers": {
"invompt": {
"command": "bunx",
"args": ["invompt-mcp"],
"env": {
"INVOMPT_API_KEY": "inv_sk_your_key_here"
}
}
}
}Requirements
Node.js 18+
An Invompt API key for
create_invoice
The resource works without INVOMPT_API_KEY. The tool requires it when called.
Environment variables
INVOMPT_API_KEY
Required for:
create_invoice
Not required for:
invompt://spec/iml/v1draft_invoice_imlif you only want draft YAML and do not call the tool
INVOMPT_API_URL
Optional override for the API base URL.
Default:
https://invompt.comUse this when targeting local, staging, or preview backends.
How it works
Resource: invompt://spec/iml/v1
Reads
GET /api/v1/spec/imlReturns plain text
Uses a shared 1-hour in-memory cache
Does not require authentication
Tool: create_invoice
Input:
{
"iml": "meta:\n invoice_number: INV-001\n issue_date: 2026-03-21\n currency: USD\nitems:\n - desc: Design work\n qty: 8\n rate: 125\n",
"templateId": "professional"
}Supported templateId values:
professionalminimalmodern
Success result:
{
"invoiceId": "uuid-here",
"url": "https://invompt.com/invoice/uuid-here"
}On failure, the tool returns an MCP error result with a JSON body containing:
{
"success": false,
"error": {
"code": "ERROR_CODE",
"message": "Human-readable description"
}
}Prompt: draft_invoice_iml
Fetches the same IML spec used by the resource
Injects it into prompt context
Instructs the model to return only raw IML YAML
Local development
Install dependencies:
npm installBuild:
npm run buildRun the server directly:
node dist/index.jsRun from source during development:
npm run devLocal package-manager verification
For unpublished local verification, register the package with both package managers:
npm link
bun linkThen verify the server through an MCP client using:
npx invompt-mcpand:
bunx invompt-mcpVerification
Verified locally:
npm run buildnpm testdirect stdio startup via
node dist/index.jsnpx invompt-mcpafternpm linkbunx invompt-mcpafterbun link
Current verification scope:
local MCP behavior is verified against a mock Invompt API
live API verification still depends on deployed WS1 routes and migration state
Publish
The package includes a GitHub Actions workflow at .github/workflows/publish.yml.
It runs:
npm cinpm run buildnpm testnpm publish --provenance --access public
License
MIT
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Tools
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Invompt/mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server