Skip to main content
Inspect workspace brands, watched brands, credits, auto top-up, monthly budget, spend by surface/kind, grants, allowances, tracked items and monthly burn, unread tracks, active keys, webhook/data-sharing status, and top-up link. Pass view: “receipt” to audit agent tool calls (args, cost, snapshot, rows — never replies) over a time window. Read-only, free.

Who may call it, and what it costs

Price. Free. Every reply says what the call cost in meta.costCredits. The prices are on Spending, tracking and stopping.

Arguments

one of account, receipt
account (default, no other argument) or receipt
string
receipt only: ISO date or time the receipt starts at; default 24 hours ago. The log keeps 30 days
string
receipt only: ISO date or time it ends before; default now
one of this, all
receipt only: this: only the calling key’s calls; all (default): every key and sign-in of the workspace
integer
receipt only: calls listed, newest first; default 100. Totals always cover the whole range min 1, max 500
string
receipt only: brd_ id of a brand of this workspace: only the calls made for it (a key bound to it, or a call that named it and no other)
string
Every argument the tool takes, as GET /v1/tools serves it, every shared object written out; https://agent.parse.gl/openapi.json serves the same schema with the filter and the window named once under components and pointed to by $ref. MCP tools/list leaves out the arguments few calls send, and says the filter and the window once, under search_answers; resolve_ids {"kinds": ["argument"]} lists what it left out, and a call with any of them runs:

Example call

Credits, the budget, spend, keys and tracked items. Over REST:
Over MCP, the assistant calls the tool by name with the same arguments:
More calls:
  • What this key’s calls read and cost, newest first: {"view":"receipt","key":"this","limit":20}
The reply has data, context, next, page and meta; follow next for the calls that come after. A refused call returns { code, message, fix }: see the error table.