Niches · Agent wallets & spending controls

Pay at an ordinary card checkout with a scoped card

Two or three companies' docs show a full fit.

2companies, full fit
2companies, fit with a limit
30companies carry the tag
18share of the category's work
4first-hand accounts

What counts as done

A single-use or merchant-locked card (or a token for the person's card) is issued for no more than the order amount, the agent completes the checkout or hands a person a link, and the card details never enter the agent's logs.

How agents ask for it: buy this with a card that only works here, make a one-time card for this order, pay for the hotel with my saved card

Does the job 2

The docs show it meets the test above, and nothing we read says otherwise.

ListingWhy
Agentcard C
Agentcard Corporation
Issuing creates a single-use card for the amount, and the Vault path has the agent submit a placeholder card while the user approves merchant and amount by passkey, so the real number never reaches the agent; approvals expire after 15 minutes and declines come back by webhook. source
Agent Vault by Self D
SocialConnect Labs, Inc., doing business as Self Labs
The agent fills a mock token and the proxy types the holder's real number for one authorised fill, under the agent card's per-transaction cap and site rules; 3-D Secure, CAPTCHA and blocks go to request_challenge_handoff. A first fill on a new site waits up to 15 minutes for approval. source

Does the job, with a limit 2

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
FluxA AI Wallet E
FluxA Technology Pte. Ltd.
Prepaid agent cards are issued for an amount under a mandate, with 3-D Secure codes readable by the CLI. card details returns the full number and CVV to the agent, identity verification is required, and holder creation was rejected by the issuer in an open issue (acc-16). source
Sponge Wallet F
Sponge Inc.
The README lists Sponge Card, vaulted-card and per-transaction virtual-card workflows and a card status command; how card details reach a checkout and the amount limits were not readable. source

Couldn't check 2

The docs were closed to us, or silent. These carry no mark either way.

ListingWhy
Kite Agent Passport D
Datalego Inc.
Scoped cards are named in the listing context, but neither the CLI reference nor the changelog read describes issuing one or how card details reach a checkout.
Privy Wallets (server wallets, agent wallets, policy engine) BB
Horkos, LLC
The agent wallets page lists card spend, but no card issuing or card-secrecy page was opened.

Doesn't do this job 5

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

ListingWhy
Circle Wallets (Agent Wallets, Programmable Wallets) BB
Circle Technology Services, LLC
No cards; USDC wallets only. source
Coinbase Developer Platform (Agentic Wallet, AgentKit, CDP MCP) BB
Coinbase, Inc.
No single-use or merchant-locked card; card products appear only as a third-party plugin (Virtuals). source
Openfort BB
Alamas Labs Inc.
No cards. source
Turnkey Agentic Wallets BB
Turnkey Global, Inc.
No cards. source
Yield.xyz MCP server BB
Omni Ltd. (Yield.xyz)
No cards. source

What stops agents finishing

  • merchant anti-bot checks and account creation stop the agent before payment
  • 3D Secure or a passkey step needs the person
  • card issuers do not open agent card programmes to individual developers

Around it

Before: commerce.find-product, agent-wallets.fund-and-cap. After: commerce.track-order, checkout-protocols.dispute-evidence.

Where this comes from

Weight 18 of 100 in Agent wallets & spending controls: 4 of 20 accounts (acc-01, acc-02, acc-08, acc-16); card checkout is the most asked-for job outside crypto in the HN threads

The same as data: /api/v1/use-cases/agent-wallets.card-checkout.json, or the MCP tool get_use_case with id agent-wallets.card-checkout. Other jobs in this category: /api/v1/niches?category=agent-wallets.

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.