For teams whose people already work with AI agents

The CRM your team's agents can use. And you control.

Every employee brings their own Claude Code, Codex or Cursor. Each one gets its own token and its own rules. Every change an agent makes is previewed, approved when it has to be, and can be undone with one click.

  • Bring your own AI, we never bill for it
  • Set up from 7 questions
  • Privacy defaults for EU data
Northwind Studio · demo workspace
The Changes inbox: two changes proposed by agents wait for approval, each with a line-by-line preview
Real product, fictional demo data. No sound.
64%

of full-time workers in a Clutch survey have built their own AI agent.

Clutch, Sept 2026
91%

of those employee-built agents can reach company data, customer records included.

Clutch, Sept 2026
52%

of employees admit using AI tools without approval, often from personal accounts.

Okta, AI Agents at Work 2026

How it works

From seven questions to a CRM your agents can work in.

1

Answer seven questions

Who you sell to, your industry, how customers find you, what you track. Click cards, pick from lists, or type your own when nothing fits. Your answers become objects, fields and pipeline stages for your business.

  • 12 industries with suggested stages, all editable
  • "Other" turns into your own fields and objects
  • The whole setup is one change you preview first
Setup questionnaire with clickable industry cards
2

Give every agent its own key

Maja's Cursor, Lars's ChatGPT connector, your Claude Code: each gets a token and a rule set. Read only, edit records alone, or ask before touching the structure. Revoke one token without breaking the others.

  • Per agent: never, ask me, or alone
  • Separate rules for records, schema and deletes
  • See what each agent applied, waited for and reverted
Agents page with three agents, each with its own rules
3

Approve, apply, undo

When an agent wants a new field or a new view, it lands in Changes with a line-by-line preview. Nothing is written until you approve. Anything applied, by a person or an agent, can be reverted.

  • Dry run before every write
  • Revert with a conflict check
  • Deletes are soft, always
A pending change from an agent with Approve and Reject

Control

You decide what agents see, and what they may do alone.

Fields agents never see

Mark a field private and it never leaves the server over MCP. Agents cannot read, search or write it, whatever model they run. Only a person can change that.

Every call, logged

Each search, read and proposal is in the activity log with the agent's name. You see what your agents look at, not only what they change.

One-click undo

Every change stores its inverse. Revert a whole import, a renamed stage or a deleted record, and see if someone changed it since.

Privacy defaults from your answers

Data must stay in the EU and someone uses a model hosted outside it? Email and phone are hidden from agents by default, and the setup tells you why.

Agent privacy settings with a visibility toggle per field
Agent privacy, per field
Activity log listing every agent call
Activity, per agent

Your AI

Any agent that speaks MCP. Your model, your tokens.

Other CRMs sell you their assistant and bill you per credit. Here your team keeps the tools they already pay for. The CRM never calls a model itself, so there is nothing to bill.

  • Claude Code
  • Claude Desktop
  • OpenAI Codex
  • ChatGPT connectors
  • Cursor
  • Gemini CLI
  • Local models via Ollama
  • Your own agent on n8n
Connect Claude Code
claude mcp add --transport http agent-crm \
  https://your-workspace/mcp \
  --header "Authorization: Bearer acrm_…"

The Agents page gives you this line with a real token, plus the same for Codex and Cursor.

Daily use

Calm enough to work in all day.

Tables and a pipeline built from your own data model. Light or dark, six accent colors, three densities. Drag the handle.

Pipeline in light mode
Pipeline in dark mode

The first test

We are looking for 10 teams.

Agent CRM is in its first phase. It works, it is tested, and it has not met enough real teams yet. That is what this test is for.

What you get

  • Free use during the test
  • We set up your workspace with you
  • Your feature requests shape what comes next

What we ask

  • At least one person who uses an AI agent at work
  • Two short feedback calls in four weeks
  • Honest answers, especially the critical ones

Join the test

AI tools your team uses

We store your answers to contact you about the test, nothing else. No newsletter. Privacy

FAQ

Questions teams ask first.

Do you train on our data or send it to a model?

No. The CRM never calls a model. Your agents call the CRM over MCP, with your model and your tokens. Fields you mark private never leave the server.

Where is our data stored?

During the test we agree the setup with each team. The plan for general use is hosting in the EU. If your data has to stay in one country, tell us in the form.

What does it cost?

Nothing during the test. After it, a flat price per workspace. AI is never billed, because you bring your own.

Which agents work with it?

Anything that speaks the Model Context Protocol over HTTP: Claude Code and Desktop, OpenAI Codex, Cursor, Gemini CLI, local models and your own agents. Each one gets its own token.

What if an agent makes a mess?

Its change shows up in Changes with a preview. Changes to the structure and all deletes wait for approval by default. Anything applied can be reverted, and you can revoke the agent's token in one click.

Can we import from HubSpot or a spreadsheet?

Today your agent can move records in through a reviewed change. A one-click CSV import is next on the roadmap.