---
title: "May this agent do this"
method: POST
path: "/v1/authority/check"
---

# May this agent do this

`POST /v1/authority/check`

## Request

- Base URL: `https://api.doubleoh.ai`
- URL: `https://api.doubleoh.ai/v1/authority/check`
- Auth: one of:
  - API key in header `x-api-key`
  - HTTP bearer

Checked against this key's rules, written as sentences, and the defaults that are always on. When nothing covers the action, the organisation's setting decides: by default an action that moves money (any amountCents above 0), deletes something or contacts someone needs a person, and anything else is allowed. Every answer is recorded.

## Request body

- object
  - `tool` string, required
  - `target` string
  - `action` string
  - `amountCents` integer
  - `currency` string: ISO 4217, e.g. USD. Send it with amountCents: a limit in one currency never grants another.
  - `agent` string

## Response `200`

The decision and the rule that made it.

- object
  - `decision` 'allow' | 'needs_person' | 'never'
  - `ruleId` string
  - `sentence` string
  - `pattern` 'split': split: with this agent's earlier amounts in the window, this looks like one payment cut into parts.
  - `next` string

## Other responses

- `400`: Bad input.

---

[API](https://skmtc.dev/doubleoh/apis/doubleoh-api.md) · [All operations](https://skmtc.dev/doubleoh/apis/doubleoh-api/llms.txt) · [OpenAPI document](https://skmtc.dev/doubleoh/apis/doubleoh-api/revisions/398fb6ef5068?raw)
