Browse workflows

Danish sales and finance teams

Prospect company brief with CVR, CRM, and email

Resolve a Danish prospect by CVR or name, enrich with CRM context, and draft a follow-up email.

Outcome: A concise company brief with registry facts, CRM history, and a draft outreach message

Integrations in this workflow

Example questions

Try asking your AI client

  • Look up CVR 25052943 (or search by company name), summarize legal form, address, industry, and key officers. Cross-check HubSpot for an existing company or contact. Draft a short Gmail follow-up referencing the registry status and any open CRM deals.

  • Jack's company just emailed me. Pull the CVR record, see if they are in HubSpot, check Dinero for open invoices, and draft a Gmail reply - do not send.

  • Which HubSpot companies have a CVR number that looks inactive, and do we still have unpaid Dinero invoices or unanswered Gmail?

  • Write a one-page prospect brief: CVR facts, HubSpot stage, last Gmail, and Dinero balance. Do not email the customer.

  • Find officers on the CVR record, match them to HubSpot contacts, and draft a Gmail intro that mentions any open Dinero invoices. Do not send.

The problem

Sales teams manually cross-check CVR, CRM, and email when qualifying Danish prospects. Registry facts and CRM history rarely sit in one place for AI agents.

Gateway tools

CVR (public registry):

  • cvr__search_companies_by_name → cvr__get_company_by_cvr
  • cvr__list_company_participants
  • cvr__list_production_units

CRM (pick one if connected):

  • HubSpot: hubspot__search_companies, hubspot__search_contacts

Email:

  • gmail__search-messages-tool or gmail__list-threads-tool

Optional finance context:

  • dinero__list_contacts, dinero__list_invoices

Governance

  • CVR returns public registry data only — still review participant names before external sharing.
  • CRM and email tools may contain personal data; scope API tokens to sales stakeholders and audit tool calls weekly.

Connect via private gateway

Activate the integrations above, create an API token scoped to mcp:private-gateway, then add this config to your MCP client.

Config file: ~/.cursor/mcp.json

{
    "mcpServers": {
        "stackgate-private-gateway": {
            "url": "https://stackgate.ai/mcp/private",
            "headers": {
                "Authorization": "Bearer YOUR_API_TOKEN"
            }
        }
    }
}