Sign inBook a demo
AI and data teamsComing soon

Let your agent make decisions your business can rely on. Keep your agent and your model.

Your agent calls Mondian's decision tools over MCP. Each decision is calculated from your data and rules, and comes back with its record.

Connect your agentBook a demo
// your agent callsreorder_stock({ "category": "Garden furniture", "week": 43 }) // Mondian returns{ "decision": "117 stores, 41,600 units", "status": "waiting_for_approval", "reason": "38 stores are above 500 units", "record": "rec_…"}
Decision recordExample
Reorder stock · Garden furniture · week 43
Made by
Your agent · Reorder stock
Approval
Waiting for Category planner
Data
sales_weekly v4 · stock 06:00
Rules
Approval above 500 units per store
Works with any agent that supports MCP
ClaudeCodexChatGPTCursorYour own agent
Why add Mondian

Your agent keeps its interface. Its decisions become calculated and recorded.

01

Decisions, not guesses.

Your agent stops improvising calculations. Each decision runs as code through a tested tool, outside the model.

Same question, three runsExample
Units to reorder · Store 12 · garden furniture · week 43
Prompt only380520430
Mondian tool410410410
02

Your rules, every call.

Limits and approval steps are applied by the tool, not left to the prompt.

Ask MondianExample
Approving a new decision tool

Reorder stock is ready and passed every check. Approve it to go live?

Independent review passed
Replayed on the last 12 weeks
Rules held: 3 of 3
Approve and go liveReject
How it works

Three steps. Then every call works the same way.

  1. 01Sign up

    Create your Mondian workspace. You get the address your agent connects to.

  2. 02Add Mondian to your agent

    One address in Claude, Codex, ChatGPT, Cursor or your own agent.

  3. 03Call a decision tool

    Your agent sees every decision your workspace has, with its inputs and what it returns.

See the setup for Claude, Codex and others

Your data is never used to train models. It is processed in an isolated sandbox and deleted afterwards, or Mondian runs inside your own systems.

Questions

Before you connect.

Which agents can connect?

Any agent that supports MCP, including Claude, Codex, ChatGPT and Cursor, and agents you build yourself.

Which model does it need?

Any. The decision runs outside the model, so changing the model your agent uses does not change the decision.

Isn't a skill enough?

A skill is text the model reads, and it may or may not follow it. A Mondian decision tool runs outside the model, so the calculation and the rules are applied every time.

Do we still need the Mondian agent?

No. Use your own agent, the Mondian agent, or both. They call the same decision tools and write the same records.

Where does our data go?

Your data is never used to train models. It is processed in an isolated sandbox and deleted afterwards, or Mondian runs inside your own systems.

Bring one decision your agent should make.

Connect your agent today, or book a 30-minute call and we connect it with you.

Connect your agentBook a demo
Film