AI and memory

aiReview.evaluate

Manually dry-run or save a deterministic and human-evidence evaluation for up to 50 redacted chats over at most 31 days. Cost limit must be zero; no model calls, schedules or production prompt changes. Requires evaluate and read grants for each exact owner. Human dashboard or terminal only.

POST/api/v1/accounting/aiReview.evaluate

Permissions

accounting:readaccounting:write

Who can call it

admin, finance, contributor

Retries

No idempotency key

MCP

Not offered over MCP: Private redacted chats and human review stay in Oatmilk and its terminal; connected AI apps receive no reviewer access or chat sharing.

Fields

  • selectionobjectRequired

    No other fields

  • dryRunboolean
  • maxCostUsd0

    Default 0

  • evaluatorVersion"deterministic-human-v1"

    Default "deterministic-human-v1"

Example

curl https://app.getoatmilk.com/api/v1/accounting/aiReview.evaluate \
  -H "Authorization: Bearer $OATMILK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "selection": {
    "from": "2026-09-30T14:00:00Z",
    "to": "2026-09-30T14:00:00Z"
  }
}'
Response
{
  "data": { … }
}

This action checks your data too, so replace the example values with your own before sending it.

Try it

Try it

Checks your input with this action’s real schema and answers like the API, with synthetic data. No key needed, and nothing changes.

POST/api/v1/accounting/aiReview.evaluate
curl https://app.getoatmilk.com/api/v1/accounting/aiReview.evaluate \
  -H "Authorization: Bearer $OATMILK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "selection": {
    "from": "2026-09-30T14:00:00Z",
    "to": "2026-09-30T14:00:00Z"
  }
}'

More in AI and memory.