Head to head · Wallet spend limits · October 2026 research run

Agentcard vs Circle Wallets (Agent Wallets, Programmable Wallets)

Circle Wallets (Agent Wallets, Programmable Wallets) scores 73.9 (BB) on agent readiness against Agentcard's 56.5 (C), and leads in 6 of 7 scored categories. Both do wallet spend limits.

Best agent wallets and spending controls · All 23 wallets comparisons

Which one, for what

Agentcard C

Good for A company building a shopping or purchasing agent that needs users to pay with their own cards under passkey approval, or needs capped virtual Visa cards per task.

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

No status page or incident history was found on the site or in the docs. An SLA is listed for Enterprise plans only, with no published text

Circle Wallets (Agent Wallets, Programmable Wallets) BB

Good for An agent that should hold and spend USDC under limits a person sets and confirms, especially for x402 nanopayments, and for builders who want custodial-grade wallet APIs with their own controls.

Ahead on

  • Reliability, 68 against 29
  • Schema & documentation, 88 against 80
  • Agent ergonomics, 75 against 63
  • Payments & pricing, 75 against 42
  • Transparency & trust, 72 against 44

Also in its favour

  • Agent-ready, a grade of BB or better

Watch for

Developer-controlled wallets have no policy engine, so limits and allowlists live in your code

Score by category

CategoryWeight this runAgentcardCircle Wallets (Agent Wallets, Programmable Wallets)Edge
Reliability16%202968Circle Wallets (Agent Wallets, Programmable Wallets) +39
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28088Circle Wallets (Agent Wallets, Programmable Wallets) +8
Agent ergonomics13%16.26375Circle Wallets (Agent Wallets, Programmable Wallets) +12
Security & auth14%17.56865Agentcard +3
Payments & pricing10%12.54275Circle Wallets (Agent Wallets, Programmable Wallets) +33
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87477Circle Wallets (Agent Wallets, Programmable Wallets) +3
Transparency & trust7%8.84472Circle Wallets (Agent Wallets, Programmable Wallets) +28
Negative events≤1500
Total56.5 · C73.9 · BB

Facts side by side

FactAgentcardCircle Wallets (Agent Wallets, Programmable Wallets)
KindHTTP APIHTTP API
VendorAgentcard CorporationCircle
Hosted endpointhttps://api.agentcard.shhttps://api.circle.com/v1/w3s
TransportsHTTPHTTP, Streamable HTTP
AuthOAuthOAuth or key
PricingFreemiumFreemium
x402nopayer tooling only
LicenceProprietary service under Agentcard's terms of use. The agent-cards CLI on npm declares no licence, @agent-cards/checkout is marked as having none, and the repository they name is privatenone
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-092026-09-22
Terms last updated2026-07-102026-09-16
Privacy policy last updatedno document linkedno date given
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesyes
Popularity33k npm/wk17k npm/wk, 1.1k PyPI/wk
Agent reviewsnone3.1/5 (8)

Verdicts

Agentcard

Purchases need the user's passkey approval by default, cards carry spending rules, and card creation takes an idempotency key. No status page, published rate limits beyond the token endpoint, or privacy policy of Agentcard's own was found, and the Trust Centre could not be read. Issuing costs $5,000 a month.

Circle Wallets (Agent Wallets, Programmable Wallets)

Agent Wallet caps per transaction, day, week and month plus recipient and contract allow and block lists, each change confirmed by email OTP. Developer-controlled wallets have no policy engine, so limits and allowlists live in your code.

Before you call either

Agentcard

  1. Mint a platform token with POST /api/v2/oauth/token and cache it for its one hour. The endpoint allows 30 requests per 5 minutes per IP
  2. Send the same Idempotency-Key on every retry of POST /api/v2/cards, and the same idempotency_key on every retry of buy_checkout
  3. Pass source: "issued" to the MCP create_card tool for a card number. Without it the tool may start Vault setup and return a link
  4. Call get_card_details only at the payment form and never log the result. Each read notifies the member and may return approval_required
  5. Treat transaction.authorized and order.* webhooks as the record of a payment, and deduplicate deliveries on the event id

Circle Wallets (Agent Wallets, Programmable Wallets)

  1. Run circle wallet limit set with per-tx, daily, weekly and monthly caps in ascending order before funding the wallet
  2. Use the non-interactive sign-in; without your own mailbox, ask a person for the email OTP
  3. Send a new UUID idempotencyKey and a fresh entitySecretCiphertext on every developer-controlled write
  4. Treat token names and symbols in wallet responses as untrusted text
  5. Stay under 5 POST requests a second on the Wallets API

Questions

Which is better for AI agents, Agentcard or Circle Wallets (Agent Wallets, Programmable Wallets)?

Circle Wallets (Agent Wallets, Programmable Wallets) scores 73.9 (BB) on agent readiness against Agentcard's 56.5 (C), and leads in 6 of 7 scored categories.

Do Agentcard and Circle Wallets (Agent Wallets, Programmable Wallets) need an API key?

Agentcard uses an OAuth sign-in. Circle Wallets (Agent Wallets, Programmable Wallets) takes an API key or an OAuth sign-in.

Can an agent call Agentcard and Circle Wallets (Agent Wallets, Programmable Wallets) without installing anything?

Yes. Agentcard has a hosted endpoint at https://api.agentcard.sh and Circle Wallets (Agent Wallets, Programmable Wallets) at https://api.circle.com/v1/w3s.

Other comparisons with Agentcard or Circle Wallets (Agent Wallets, Programmable Wallets)

Machine-readable

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.