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 add spend controls to an n8n AI agent workflow

1. Why spend control matters in n8n

n8n makes it fast to wire together AI models, APIs and data sources into automated workflows. Once those workflows start calling paid AI services — OpenAI, Anthropic, data vendors — there is no native gate between the workflow trigger and the money leaving your account.

NORNR adds a decision node: the workflow asks whether a spend is permitted before the paid step runs. Low-risk calls continue automatically. Higher-value calls queue for operator review. Out-of-policy calls are blocked with a clear reason.

2. Install the SDK

bash

pip install agentpay

The SDK is a thin wrapper around the NORNR policy engine. It submits a mandate check and returns a decision object before your n8n Code node proceeds.

3. Create the governed wallet

In an n8n Code node that runs before your paid step, create a wallet with a daily budget and approval threshold:

python

from agentpay import Wallet

wallet = Wallet.create(
    owner="n8n-research-workflow",
    daily_limit=50,
    require_approval_above=15,
    base_url="https://nornr.com",
)

4. Gate the paid step

Before the AI API call node runs, add a Code node that requests a decision:

python

decision = wallet.pay(
    amount=8.00,
    to="openai",
    purpose="document summarisation",
)

if decision.get("status") == "approved":
    # continue to the AI node
    return [{"json": {"proceed": True, "decision": decision}}]
elif decision.get("status") == "queued":
    # halt workflow, notify operator
    return [{"json": {"proceed": False, "reason": "queued for review", "decision": decision}}]
else:
    # blocked — log and stop
    return [{"json": {"proceed": False, "reason": "blocked by policy", "decision": decision}}]

Add an n8n IF node after this Code node that checks proceed === true. The true branch continues to the AI step; the false branch routes to a notification or dead-letter node.

5. What the decision object looks like

output

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

6. Common n8n mistakes

Placing the NORNR check after the AI API node instead of before it.
Using a single hard-coded amount — estimate conservatively if the real cost varies.
Not handling the queued path — treat it as a pause, not a failure.