List activity - Interchange API

GET /api/v2/storefront/audit-logs Returns the storefront activity feed: recent configuration and inventory-source changes, with the actor and before/after values attributed per row. Filter by resource type, inventory source, and date range.

Request

curl

curl -G https://api.interchange.io/api/v2/storefront/audit-logs \
  -H "Authorization: Bearer $SCOPE3_API_KEY" \
  --data-urlencode "resourceTypes=STOREFRONT_INVENTORY_SOURCE" \
  --data-urlencode "inventorySourceId=source_abc123" \
  --data-urlencode "startDate=2026-03-01T00:00:00Z" \
  --data-urlencode "take=50"

Parameters

Field	Type	Required	Notes
`startDate`	string	No	ISO timestamp; include logs at or after this time
`endDate`	string	No	ISO timestamp; include logs at or before this time
`inventorySourceId`	string	No	Filter to a single storefront inventory source by source ID
`resourceTypes`	enum \| array	No	`STOREFRONT` or `STOREFRONT_INVENTORY_SOURCE`. Defaults to all storefront resource types
`actorOrigin`	enum	No	`human`, `agent`, `autonomous`, or `system`. `agent` surfaces only changes Murph made on a user’s behalf; `human` only direct user changes
`outcome`	enum	No	`succeeded` (the default feed), `denied`, or `failed`. Omit to see only changes that took effect; pass `denied` for blocked attempts or `failed` for errored ones. The `denied`/`failed` views are not limited to the default resource types
`actorUserId`	integer	No	Filter to actions taken by a specific user (by user id)
`actorUserEmail`	string	No	Filter to actions taken by a specific user (by email)
`resourceId`	string	No	Filter to a single resource by its external id (e.g. an inventory source or advertiser id)
`actions`	enum \| array	No	Narrow to a subset of meaningful actions (e.g. only `DELETE`/`ARCHIVE`). Defaults to all meaningful actions
`take`	integer	No	Results to return, max `500` (default `50`)
`skip`	integer	No	Results to skip for pagination (default `0`)

Response

{
  "logs": [
    {
      "id": 90217,
      "timestamp": "2026-03-04T18:22:11Z",
      "createdAt": "2026-03-04T18:22:11Z",
      "action": "UPDATE",
      "resourceType": "STOREFRONT_INVENTORY_SOURCE",
      "resourceId": "source_abc123",
      "resourceName": "Northwind CTV",
      "userId": 4412,
      "userEmail": "ops@northwind.example",
      "userName": "Northwind Ops",
      "serviceTokenId": null,
      "serviceTokenName": null,
      "actorOrigin": "human",
      "outcome": "succeeded",
      "parameters": { "status": "ACTIVE" },
      "changes": { "status": { "from": "PENDING", "to": "ACTIVE" } },
      "description": "Activated inventory source Northwind CTV"
    }
  ],
  "total": 1
}

Field	Type	Notes
`logs[]`	array	Audit rows, newest first
`logs[].action`	string	The change applied (e.g. `CREATE`, `UPDATE`, `DELETE`)
`logs[].resourceType`	string	`STOREFRONT` or `STOREFRONT_INVENTORY_SOURCE`
`logs[].resourceId` / `resourceName`	string \| null	The affected resource
`logs[].userId` / `userEmail` / `userName`	mixed \| null	Actor when a user made the change
`logs[].serviceTokenId` / `serviceTokenName`	mixed \| null	Actor when a service token made the change
`logs[].actorOrigin`	string \| null	How the action was initiated: `human`, `agent` (Murph acting on a user’s behalf), `autonomous`, or `system`
`logs[].outcome`	string \| null	Whether the action took effect: `succeeded`, `denied` (blocked by policy/permission), or `failed` (errored mid-flight)
`logs[].parameters`	object \| null	Request parameters captured for the action
`logs[].changes`	object \| null	Before/after values
`logs[].description`	string	Human-readable summary
`total`	integer	Total matching rows (for pagination)

Who acted, and whether it took effect

Two fields answer the questions operators ask of any change:

actorOrigin — was it a person, the agent, automation, or the platform? agent means Murph made the change on a user’s behalf; human is a direct user change. Filter with ?actorOrigin=agent to see only what the agent did.
outcome — did it actually happen? The default feed returns only succeeded changes. To audit what was blocked or failed, pass ?outcome=denied (blocked by policy or permission) or ?outcome=failed (errored mid-flight). These attempt views intentionally widen beyond the default resource types so a denied attempt on any resource surfaces.

Connecting a payout account is recorded as a REGISTER action and is included in the default feed, so “who connected our payout account” is answerable here.

Operators don’t have to call this endpoint directly. In the storefront chat, opening Activity log (or /activity) renders this feed as a read-only panel with actor-origin and outcome chips, and asking Murph a specific question — “who changed our billing?”, “what changed yesterday?”, “what got blocked?” — answers it in prose from the same feed.

Errors

400 VALIDATION_ERROR — malformed date or take over 500.
401 UNAUTHORIZED — missing or invalid API key.

See Errors for the full error contract.

Activity tasks

All activity operations

Inventory sources

The sources these logs track

Get reporting metrics

Delivery metrics

​Request

​Parameters

​Response

​Who acted, and whether it took effect

​Errors

​Related

Activity tasks

Inventory sources

Get reporting metrics

Request

Parameters

Response

Who acted, and whether it took effect

Errors

Related