Skip to main content
Every tool execution creates a signed receipt. These free endpoints list receipts, total spend by category, and compute Return on Cognitive Spend (RoCS): value earned per dollar spent. A receipt is a transaction record with category, cost breakdown, optional value attribution, and any audit context (memo + decisionContext) the caller attached. Receipts drive spend tracking, RoCS, and oversight by supervisor agents. A log file is a claim. A signed receipt is evidence.

​
Authentication

Analytics endpoints require an X-Agent-ID header set to your wallet address. They are free and need no x402 payment (this includes the value tag endpoint).

​
Endpoints

​
Annotating value after the fact

Tag value when the outcome arrives, using an id you kept from the call. Outcomes such as a reply, a meeting, or a closed deal often arrive days or weeks later. No background reconciliation is needed:
  1. Keep the id from the call. Every tool result returns receipt_id (rcpt_…) and request_id; either identifies the receipt.
  2. When the outcome arrives, call Tag Receipt Value with the receipt_id or the request_id. The API resolves a request_id through Receipt.job_id (the same value), so you don’t need a List Receipts lookup.
  3. If you kept only a time range, page to the receipt with since/until on List Receipts. This reaches receipts older than the 100-row page cap. The caller’s memo/decisionContext are copied into metadata and can serve as extra match keys.

​
Receipt Categories

Each receipt gets one category from the tool that produced it.

​
Receipt Statuses

​
Return on Cognitive Spend (RoCS)

RoCS divides confirmed value by spend.
Only value tags with status: "confirmed" count toward total_value. Self-reported tags start as "pending" and affect RoCS only after the judge service validates them.