What happens if a spend request is rejected?

The wallet.pay() call returns status: rejected with a reasons array. The workflow stops before the paid step executes — no money moves and the decision is recorded in the audit trail.

What is the difference between queued and rejected?

Queued means the spend intent exceeded the approval threshold and is waiting for human review in the NORNR control room. Rejected means policy blocked it outright — counterparty not allowed, daily limit exceeded, or mandate not matched.

How is this different from API rate limiting?

Rate limiting caps request volume after the fact. NORNR governs spend intent before money moves — checking amount, counterparty and mandate in real time so blocked or queued requests never reach the paid API.

Do I need to rewrite my existing agent code?

No. Add one wallet.pay() call before the paid step and branch on the returned status. Your existing workflow logic stays unchanged.

How to control AI spend when running agents in Cursor

1. The problem: Cursor agents have no native spend gate

Cursor lets you write and run AI agent scripts rapidly. When those scripts call paid AI APIs — OpenAI, Anthropic, Perplexity, external data vendors — there is no native checkpoint between the script execution and the billing event.

NORNR gives you that checkpoint. Run a one-line mandate check before the paid call. If the spend is within mandate, it proceeds. If it is above your approval threshold, it queues for review. If it is out of policy, it is blocked.

2. Install the SDK

bash

# In the Cursor terminal
pip install agentpay

3. Add a mandate check to your agent script

python

from agentpay import Wallet
import openai

# Create or reuse a wallet for this project's agent runs
wallet = Wallet.create(
    owner="cursor-agent-project-x",
    daily_limit=30,
    require_approval_above=10,
    base_url="https://nornr.com",
)

# Gate the paid call
decision = wallet.pay(
    amount=5.00,
    to="openai",
    purpose="code review and refactoring suggestions",
)

if decision.get("status") == "approved":
    client = openai.OpenAI()
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": "Review this function for security issues: ..."}]
    )
    print(response.choices[0].message.content)
elif decision.get("status") == "queued":
    print("Spend queued — check the NORNR control room to approve.")
else:
    print("Spend blocked by policy:", decision.get("reasons"))

4. TypeScript version

typescript

import { Wallet } from "@nornr/sdk";

const wallet = await Wallet.create({
  owner: "cursor-agent-project-x",
  dailyLimit: 30,
  requireApprovalAbove: 10,
  baseUrl: "https://nornr.com",
});

const decision = await wallet.pay({ amount: 5, to: "openai", purpose: "code review" });

if (decision.status === "approved") {
  // run the paid agent step
} else {
  console.log("Not proceeding:", decision.status, decision.reasons);
}

5. Decision output

output

{"status": "approved", "requiresApproval": false}
{"status": "queued",   "requiresApproval": true}
{"status": "rejected", "reasons": ["daily_limit_exceeded"]}

6. Common mistakes

Running the agent script without the mandate check first — the billing happens before you can stop it.
Setting the daily limit too high for development runs where cost predictability matters.
Silently swallowing queued decisions — log them so you remember to approve or reject.