Niches · Pay-per-call protocols

Check the price and payee and stay inside a budget before paying

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

0companies, full fit
3companies, fit with a limit
24companies carry the tag
18share of the category's work
4first-hand accounts

What counts as done

The client refuses to sign when amount, asset or payee differ from what was expected or the task budget is spent, reports why in a machine-readable way, and keeps a record of every payment made in the task.

Measured, for our own tests: an inflated 402 (10 times the expected price) is refused in a test. No docs we read state these figures, so they don't count against a listing here.

How agents ask for it: don't spend more than a dollar on this, only pay if it costs what the docs say, set a budget for this research task, stop if the price changes

Does the job, with a limit 3

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
L402 C
Lightning Labs, Inc.
lnget refuses invoices above --max-cost (default 1000 sats) and fees above --max-fee (default 10), and the spec tells clients to abandon excessive invoices. A cap is per invoice with no per-task budget, failures are only 401 or 402, and an inflated 402 below the default cap would be paid. source
Machine Payments Protocol (MPP) A
Tempo and Stripe
The spec says clients MUST check amount, recipient, currency and validity and never trust description, and mppx sessions take a deposit ceiling. No per-payment or per-task client cap is documented for charges, and one builder found no standard budget before payment (acc-08). source
x402 A
x402, a Series of LF Projects, LLC
spendControls defaults to a 1 USD per-payment ceiling on known assets and policies can drop accepts with an unexpected payTo, with reasons like recipient_mismatch. No per-task budget exists, and a black-box test paid a server that inflated the price tenfold under the cap (acc-10, open). source

Doesn't do this job 1

Listings in the category that do something else, with the reason.

ListingWhy
ACK-Pay (Agent Commerce Kit) D
Catena Labs, Inc.
Amount, currency and recipient sit in the signed request, but the SDK has no per-payment or per-task cap, open issue 222 says expiresAt is never enforced, and error taxonomy is left to implementers. source

What stops agents finishing

  • client signs whatever amount the 402 asks for unless the caller sets a cap
  • spend budgets sit outside the spec, so each client implements its own
  • retries and rephrased queries each trigger a new paid call

Around it

Before: agent-wallets.fund-and-cap. After: pay-per-call.pay-and-retry, agent-observability.cost-report.

Where this comes from

Weight 18 of 100 in Pay-per-call protocols: 4 of 25 accounts (acc-01, acc-08, acc-09, acc-10); every buyer thread raised budgets, and the x402 overcharge test is the most discussed client bug read

The same as data: /api/v1/use-cases/pay-per-call.cap-and-check.json, or the MCP tool get_use_case with id pay-per-call.cap-and-check. Other jobs in this category: /api/v1/niches?category=pay-per-call.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.