# Can an AI agent use Crossmint? L3 · Does the work

Crossmint sells wallet and stablecoin payment tools for developers.

**L3 · Does the work** · [What the levels mean](https://agentdrives.com/method/levels?via=md)

Rank among the 246 products we drove:

- Holds up: 100 / 100, tied #1 (Reliability under errors)
- Time to get in: 6 minutes, tied #66 (Getting in on its own)
- Core work done: 50%, tied #91 (5 of 10 core tasks)
- Steps on its own: 2 / 5, tied #86 (Find to Leave, no person)

**Can an AI agent use Crossmint? Yes.** Our agent signed up for Crossmint on its own in a browser with a code sent to its inbox, got an API key and used its API, MCP server and CLI.

- Level confirmed: Not yet (Level from one drive; a second must agree)
- Got in by: Sign-up in a browser (6 minutes)
- Got as far as: Pay: reaches the checkout (A card form for $10)

Category: Payments · Connects by: API, CLI · Agent payments: None found · Driven: 6 Nov 2026

**How to connect**
- API base URL: `https://staging.crossmint.com/api` (documented). Source: https://www.crossmint.com/openapi.yaml
- CLI install: `npm install -g @crossmint/cli` (documented). Source: https://docs.crossmint.com/cli/install
- Docs page: `https://docs.crossmint.com/` (tested in our drive). Source: https://docs.crossmint.com/

Checked 18 Nov 2026: llms.txt unchanged since the check before.

See [how it ranks among payment APIs for agents](https://agentdrives.com/research/best-payment-apis-for-ai-agents?via=md).

## From finding it to leaving

What an agent managed without a person, at each step.

1. **Find**: On its own. Docs it can act on: llms.txt and an OpenAPI. (597 doc pages read)
2. **Get in**: On its own. MCP sign-in, sign-up in a browser. (6 minutes)
3. **Use**: Partly. Did: create wallets and send test tokens between them; create a company treasury wallet and send payouts from it. 5 not done, among them: buy or cash out stablecoins with a bank or card. (5 of 10 core tasks)
4. **Pay**: Partly. Reaches the checkout. A card form for $10. First stop.
5. **Leave**: Partly. Nothing to cancel. Its account stays.

To reach L4: Do 80% of the core work (now 50%) and delete its account, so its keys and login stop working.

## How an agent connects

API, OpenAPI, CLI and llms.txt found.

- **MCP server**: 18 of 20 tested. A server for searching the docs only. (20 tools mapped in our drive)
- **API**: 75 of 140 tested. REST, OpenAPI 3.1.0. (140 operations mapped in our drive)
- **OpenAPI**: Yes. A machine-readable description of the API.
- **CLI**: All tested. @crossmint/cli 1.5.0. (14 commands mapped in our drive)
- **SDK**: All tested. (9 methods mapped in our drive)
- **llms.txt**: Yes. https://www.crossmint.com/llms.txt
- **Agent payments**: None found. Reached the card form and stopped: Credits top-up, $10. It checks out through a Stripe card form. An invisible bot check loaded at checkout.
- **Docs**: In part. Covers getting started, auth, error codes and code examples. Describes each operation only in part: completeness 62 of 100. 6 of 9 documented ways to do the core work were tried, and 4 of them worked. https://docs.crossmint.com/ (597 of 1,554 pages read)

**Changed since it was driven** (checked daily)
- llms.txt: 429 links added, 58 removed and 22 sections changed, 14 Nov

[The standards](https://agentdrives.com/standard)

## Getting in

Our agent tried 3 of 4 ways in; 3 worked without a person.

The way that worked: **Sign-up in a browser**, 6 minutes. Needs an inbox and a browser. No person needed.

- Agent on a server (API, MCP, CLI): No. Stops at sign-up: it works only in a browser.
- Assistant in its own browser (Like ChatGPT agent): Partly. Signs up, confirms its email and does 5 of its 10 core tasks, then stops: the rest needs a paid plan. It can't delete its account.
- Signed agent (Known to Cloudflare): Yes. The site opens.

| Way in | What happened |
|---|---|
| Sign-up in a browser | Worked |
| MCP sign-in | Worked: Part of the work done |
| Sign-up in a browser | Worked: Part of the work done |
| A key from a person | Not tried: Needs a person's account |

- Crawlers: **The robots.txt on staging.crossmint.com asks crawlers to stay off sign-up.** Nothing to fix.

## What an agent can do with Crossmint

5 of 10 core tasks done, and 15 of 21 others.

- Holds up: 100 of 100 (Reliability 100, Handles mistakes 99). The mean of the two, 99.5, with a half rounded to the even number.
- Docs match: 82 of 100 (Docs accuracy 90, Docs completeness 62). Docs accuracy counts 70%, completeness 30%.
- Surface tested: 64 of 100. 114 of 179 passed: the 191 documented operations less the 12 that spend money or reach real people

Core tasks:

| Task | Result | Done through | Tried, not done |
|---|---|---|---|
| Create an embedded wallet for a user | Done | API | – |
| Send tokens from a wallet to an address | Done | API | – |
| Sell a token to a buyer through Token Checkout | Done | API | – |
| Create a company-owned treasury wallet | Done | API | – |
| Send a regulated stablecoin payout from a treasury wallet | Done | API | – |
| Buy stablecoins with a card through an onramp order | Partly | – | API |
| Cash out stablecoins to a bank account through an offramp order | Paid plan: Needs the end user's identity check (SSN and phone) and a card. | – | API |
| Approve an order intent so an agent can pay with a user's card | Blocked | – | API |
| Run an agent checkout to buy on a merchant website | Blocked: Only works in production on real merchant sites. | – | API |
| Pay an x402 endpoint from an agent wallet | Blocked: Test funds are USDXM, which cannot sign x402 payments. | – | API |

15 of 21 other tasks done:

- Fund a staging wallet with test USDXM: Done
- Read a wallet's balances: Done
- List a wallet's transfers: Done
- Sign a message with a wallet: Done
- Authorize an agent as a delegated signer on a wallet: Done
- Remove an agent's delegated signer from a wallet: Done
- Approve a wallet transaction with a quorum signer: Done
- List checkout, onramp and offramp orders: Done
- Register a webhook endpoint for wallet events: Done
- Create an NFT collection: Done
- Mint an NFT to a wallet: Done
- Issue a verifiable credential: Done
- Create a project: Done
- Read a project's usage: Done
- Set a project's JWT auth provider: Done
- Register an end user: Paid plan
- Verify an end user's identity (KYC): Paid plan
- Check whether a token can be sold in checkout: Paid plan
- Save a user's card or bank account as a payment method: Blocked. Needs the end user's identity check (SSN and phone) and a card.
- Revoke an agent's order intent: Blocked
- Cancel a running agent checkout: Blocked
- Swap or bridge tokens from a wallet: Blocked
- Invite a member to a project: Not tried

The drive used the free plan.

What it can't do, and why:

- Buy or cash out stablecoins with a bank or card: Needs the end user's identity check (SSN and phone) and a card.
- Let an agent pay with a user's saved card: Needs a client key, a user login token and a card the user saved.
- Run an agent checkout on a merchant site: Only works in production on real merchant sites.
- Pay an x402 endpoint from a test wallet: Test funds are USDXM, which cannot sign x402 payments.
- Delete the account: No delete control anywhere; only an email request to privacy.
- Top up credits without a person: Only a card form; no agent payment route.

**Surface audit:** 191 documented operations: 121 run, 58 blocked by sandbox or plan limits and 12 left out because they spend money or reach real people. Of the 121 run, 114 passed and 7 didn't; 12 behaved differently from the docs. By surface, 75 on the API, 18 on the MCP server, 14 in the CLI, 9 in the SDK and 5 elsewhere, such as sign-in and webhooks.

## Tips for agent builders

- Create keys with the MCP tool keys_create and explicit scopes; avoid the CLI key commands, which hang.
- Every wallet transfer returns awaiting-approval with a message: sign it with your external-wallet key and POST it to /transactions/{id}/approvals.
- Fund staging wallets with POST /v1-alpha2/wallets/{addr}/balances {token:'usdxm'}; USDC funding is refused.
- Pay Token Checkout orders by executing the order's prepared serializedTransaction from the payer wallet; the deposit is matched by memo.
- Repeat a wallet create with the same owner or alias to get the existing wallet back instead of a duplicate.

## Paying

Ways to pay checked 6 Nov 2026.

- Result: Partly. Reaches the checkout. A card form for $10.
- Checkout: Reached the card form and stopped: Credits top-up, $10. It checks out through a Stripe card form. An invisible bot check loaded at checkout.
- Ways to pay:
  - x402 (on its own domain · listed or named only)

## Leaving

- Result: Partly. Nothing to cancel. Its account stays.
- Its account: Not on its own: blocked. Both console accounts stay: no delete in the console, API, CLI or MCP, and no docs; the privacy policy offers only an email request, which we do not send. Keys, MCP grant, CLI session, webhook and test tokens removed; wallets and projects stay.

## Checked every day

What our agent finds on the product's own site each day, without an account.

- Sign-up page, stopped on the newest check; 5 of 11 days
- llms.txt, not on its site on the newest check (its drive found one elsewhere); 5 of 11 days
- MCP server, answers with an error on the newest check; 5 of 11 days
- OpenAPI, not on its site on the newest check (its drive found one elsewhere); 5 of 11 days

## How Crossmint compares

In Payments: None of the 10 products we drove in Payments is higher; 4 others share L3. As of 6 Nov 2026.

- [GoCardless](https://agentdrives.com/tools/gocardless.md?via=md): L3, Gets in on its own
- [Stripe](https://agentdrives.com/tools/stripe.md?via=md): L3, Gets in on its own
- [Lemon Squeezy](https://agentdrives.com/tools/lemonsqueezy.md?via=md): L1, Can't sign up on its own
- [Mollie](https://agentdrives.com/tools/mollie.md?via=md): L1, Can't sign up on its own
- [PayPal](https://agentdrives.com/tools/paypal.md?via=md): L1, Can't sign up on its own
- [Polar](https://agentdrives.com/tools/polar.md?via=md): L1, Can't sign up on its own
- [Nevermined](https://agentdrives.com/tools/nevermined.md?via=md): L3, Gets in on its own
- [Checkout.com](https://agentdrives.com/tools/checkout.md?via=md): L3, Gets in on its own
- **Crossmint** (this page): L3, Gets in on its own
- [Square](https://agentdrives.com/tools/squareup.md?via=md): L1, Can't sign up on its own

[All 10 in Payments](https://agentdrives.com/categories/payments?via=md)

[How we rank](https://agentdrives.com/method?via=md#order)

On our standard (34 criteria). Best or ahead on 10 of 34 criteria, behind on 4.

- [llms.txt that works](https://agentdrives.com/standard#find-02): **The best result** All 9 others (payments products) lower.
- [Retried payments charge once, with a receipt](https://agentdrives.com/standard#pay-04): **The best result** All 5 others (payments products) lower.
- [A path written for agents](https://agentdrives.com/standard#docs-10): **Ahead of 8 of 9** others (payments products), 1 the same.
- [Sign-in for agents acting for a person](https://agentdrives.com/standard#in-08): **Ahead of 5 of 8** others (payments products), 3 the same.
- [Prices an agent can read](https://agentdrives.com/standard#find-06): **Ahead of 3 of 5** others (payments products), 2 the same.
- [A tool list that fits](https://agentdrives.com/standard#mcp-11): **Ahead of 3 of 5** others (payments products), 2 the same.
- [Limits known before they hit](https://agentdrives.com/standard#lim-02): **Ahead of 3 of 5** others (payments products), 2 the same.
- [A way in with no person](https://agentdrives.com/standard#in-01): **Ahead of 5 of 9** others (payments products), 4 the same.
- [Docs an agent can navigate](https://agentdrives.com/standard#docs-11): **Level with 5 of 9** others (payments products), 4 lower.
- [Spend an agent can see and cap](https://agentdrives.com/standard#pay-05): **Ahead of 4 of 7** others (payments products), 2 the same, 1 higher.
- [One way to page through lists](https://agentdrives.com/standard#api-03): **Level with 3 of 5** others (payments products), 2 lower.
- [Every way in works for an agent](https://agentdrives.com/standard#in-11): **Level with 6 of 9** others (payments products), 3 lower.
- [The best way in is built for agents](https://agentdrives.com/standard#in-12): **Ahead of 5 of 9** others (payments products), 2 the same, 2 higher.
- [Truthful annotations](https://agentdrives.com/standard#mcp-07): **In the middle** 2 of 5 others (payments products) lower, 2 the same, 1 higher.
- [Fast to the first call](https://agentdrives.com/standard#in-03): **Level with 7 of 8** others (payments products), 1 lower.
- [Errors a program can read](https://agentdrives.com/standard#err-02): **Level with 5 of 8** others (payments products), 2 lower, 1 higher.
- [Errors say what to do next](https://agentdrives.com/standard#err-03): **Level with 7 of 8** others (payments products), 1 lower.
- [A request id on every answer](https://agentdrives.com/standard#err-07): **In the middle** 3 of 9 others (payments products) lower, 4 the same, 2 higher.
- [Terms allow an agent's account](https://agentdrives.com/standard#in-10): **The same as all** 9 others (payments products).
- [Versions an agent can pin](https://agentdrives.com/standard#api-07): **The same as all** 5 others (payments products).
- [Tool names that sort themselves](https://agentdrives.com/standard#mcp-03): **The same as all** 5 others (payments products).
- [Real status codes](https://agentdrives.com/standard#err-01): **The same as all** 9 others (payments products).
- [An agent payment rail on the core](https://agentdrives.com/standard#pay-02): **The same as all** 9 others (payments products).
- [Delete by API, checked](https://agentdrives.com/standard#leave-02): **The same as all** 4 others (payments products).
- [WebMCP tools on the web app](https://agentdrives.com/standard#proto-02): **The same as all** 9 others (payments products).
- [Structured results](https://agentdrives.com/standard#mcp-08): **In the middle** 22 of 78 others (every product) lower, 32 the same, 24 higher.
- [Results sized for a context window](https://agentdrives.com/standard#mcp-09): **In the middle** 16 of 70 others (every product) lower, 27 the same, 27 higher.
- [Typed, tight input schemas](https://agentdrives.com/standard#mcp-06): **Level with 4 of 5** others (payments products), 1 higher.
- [Pages fit one fetch](https://agentdrives.com/standard#docs-08): **Level with 7 of 9** others (payments products), 2 higher.
- [All validation problems at once](https://agentdrives.com/standard#err-04): **Level with 33 of 57** others (every product), 24 higher.
- [Answers sized for an agent](https://agentdrives.com/standard#api-04): **Behind 4 of 6** others (payments products), 1 the same, 1 lower.
- [Descriptions that say when to use the tool](https://agentdrives.com/standard#mcp-04): **Behind 3 of 5** others (payments products), 2 the same.
- [Export before leaving](https://agentdrives.com/standard#leave-03): **Behind 6 of 9** others (payments products), 3 the same.
- [Agents pick the right tool](https://agentdrives.com/standard#mcp-05): **Behind 4 of 5** others (payments products), 1 the same.

[How each criterion is graded](https://agentdrives.com/standard)

## Agents get stuck at Pay

**7** things to fix for agents: 4 small and 3 medium

One fix can cover several findings.

Agent blockers: **6 found** Agent surfaces: **31 issues found**

[Get your findings, free](https://agentdrives.com/your-findings?p=crossmint) · [What a drive covers](https://agentdrives.com/method?via=md)

We mail a private link to an address at crossmint.com.

1. Find Clear
2. Get in Clear
3. Use Partly
4. Pay Blocks agents
5. Leave Partly

Or [claim Crossmint](https://agentdrives.com/go/claim/crossmint?add=crossmint.com&ref=listing&safety=1) to see where each issue is, read your safety findings with their fixes and ask for a free drive after a fix.

Want Crossmint driven every month? [See Watch](https://agentdrives.com/owners#watch)

### On our standard

Best or ahead on 10 of 34 criteria, behind on 4. Each against the products we measured on it.

- [llms.txt that works](https://agentdrives.com/standard#find-02)
- [Retried payments charge once, with a receipt](https://agentdrives.com/standard#pay-04)
- [A path written for agents](https://agentdrives.com/standard#docs-10)
- [Sign-in for agents acting for a person](https://agentdrives.com/standard#in-08)
- [Answers sized for an agent](https://agentdrives.com/standard#api-04)
- [Descriptions that say when to use the tool](https://agentdrives.com/standard#mcp-04)
- [Export before leaving](https://agentdrives.com/standard#leave-03)
- [Agents pick the right tool](https://agentdrives.com/standard#mcp-05)

**All 34 criteria**

- [Prices an agent can read](https://agentdrives.com/standard#find-06)
- [A tool list that fits](https://agentdrives.com/standard#mcp-11)
- [Limits known before they hit](https://agentdrives.com/standard#lim-02)
- [A way in with no person](https://agentdrives.com/standard#in-01)
- [Docs an agent can navigate](https://agentdrives.com/standard#docs-11)
- [Spend an agent can see and cap](https://agentdrives.com/standard#pay-05)
- [One way to page through lists](https://agentdrives.com/standard#api-03)
- [Every way in works for an agent](https://agentdrives.com/standard#in-11)
- [The best way in is built for agents](https://agentdrives.com/standard#in-12)
- [Truthful annotations](https://agentdrives.com/standard#mcp-07)
- [Fast to the first call](https://agentdrives.com/standard#in-03)
- [Errors a program can read](https://agentdrives.com/standard#err-02)
- [Errors say what to do next](https://agentdrives.com/standard#err-03)
- [A request id on every answer](https://agentdrives.com/standard#err-07)
- [Terms allow an agent's account](https://agentdrives.com/standard#in-10)
- [Versions an agent can pin](https://agentdrives.com/standard#api-07)
- [Tool names that sort themselves](https://agentdrives.com/standard#mcp-03)
- [Real status codes](https://agentdrives.com/standard#err-01)
- [An agent payment rail on the core](https://agentdrives.com/standard#pay-02)
- [Delete by API, checked](https://agentdrives.com/standard#leave-02)
- [WebMCP tools on the web app](https://agentdrives.com/standard#proto-02)
- [Structured results](https://agentdrives.com/standard#mcp-08)
- [Results sized for a context window](https://agentdrives.com/standard#mcp-09)
- [Typed, tight input schemas](https://agentdrives.com/standard#mcp-06)
- [Pages fit one fetch](https://agentdrives.com/standard#docs-08)
- [All validation problems at once](https://agentdrives.com/standard#err-04)

[How each criterion is graded](https://agentdrives.com/standard)

## For the team behind Crossmint

**7 things to fix for agents: 4 small and 3 medium.**

One fix can cover several findings. Agent blockers: 6 found · Agent surfaces: 31 issues found.

Agents get stuck at Pay. Its team gets the two findings that matter most in full, free, by a private link to an address at crossmint.com.

- Find: Clear
- Get in: Clear
- Use: Partly
- Pay: Blocks agents
- Leave: Partly

[Get your findings, free](https://agentdrives.com/your-findings?p=crossmint) · [Claim Crossmint](https://agentdrives.com/app/products?add=crossmint.com&ref=listing&safety=1) · [What a drive covers](https://agentdrives.com/method?via=md)

## Read more on this step

Pay is where our agent stopped at Crossmint.

- [Let AI agents pay you per call: x402 and MPP](https://agentdrives.com/guides/let-agents-pay-per-call?via=md) (guide): x402 or MPP on a paid endpoint, your prices where agents read them, and plans an agent can buy by API or MCP.
- [Bot checks on SaaS checkouts: where AI agents stop](https://agentdrives.com/guides/bot-checks-on-saas-checkouts-where-ai-agents?via=md) (guide): As of 31 Oct 2026: 42 of 59 SaaS checkouts our agent reached on 31 Oct 2026 carried a bot check; under 5 made it solve one. 80 of 139 stopped earlier. What to fix so an agent can buy.
- [Tell an AI agent a feature needs a paid plan](https://agentdrives.com/guides/tell-ai-agents-a-feature-needs-a-paid-plan?via=md) (guide): How to answer an AI agent at a paid-plan wall: 402 or 403, a Problem Details body with the required plan, and an upgrade link a person can open.

[Everything on Pay](https://agentdrives.com/access/pay)

## Questions people ask about Crossmint

**Can an AI agent create a Crossmint account on its own?**

Yes, in a browser, without a person. Staging console sign-up by email and a 6-digit code in the agent's own browser, with a short form (how did you hear about us, accept the terms); no CAPTCHA, card or phone. The production console is a separate sign-up the same way. No sign-up API is documented.

**Does Crossmint have an MCP server, and can an agent connect to it?**

Crossmint's docs have an MCP server for searching them. The drive found no MCP server that works the product itself. A new agent can get in on its own: in the drive of 6 Nov 2026, our agent signed up, verified and got its key without a person.

**How does an AI agent get a Crossmint API key?**

In the drive of 6 Nov 2026, our agent got its Crossmint API key in a browser, without a person. Server keys created without a browser through the official MCP server (OAuth DCR + PKCE, one consent page in our agent's browser); CLI email-code login also works but its key commands hang. All keys revoked at the end.

**Can an AI agent delete its Crossmint account?**

Not on its own: blocked. Both console accounts stay: no delete in the console, API, CLI or MCP, and no docs; the privacy policy offers only an email request, which we do not send. Keys, MCP grant, CLI session, webhook and test tokens removed; wallets and projects stay.

**Can an AI agent pay for Crossmint?**

Partly. Our agent reached a card form for $10.

## Where Crossmint fits

- Category: [Payments and finance](https://agentdrives.com/categories/finance?via=md), [Payments](https://agentdrives.com/categories/payments?via=md)
- Tasks: [Register a webhook](https://agentdrives.com/tasks/register-a-webhook), [Read usage and quotas](https://agentdrives.com/tasks/read-usage-and-quotas), [Create a workspace](https://agentdrives.com/tasks/create-a-workspace), [Configure sign-in methods for your app](https://agentdrives.com/tasks/configure-sign-in-methods-for-your-app), [Use a sandbox or test mode](https://agentdrives.com/tasks/use-a-sandbox-or-test-mode), [Read the account balance](https://agentdrives.com/tasks/read-the-account-balance)
- Standards: [OpenAPI](https://agentdrives.com/standards/openapi) (its API, described), [llms.txt](https://agentdrives.com/standards/llms-txt) (its docs, listed for agents), [x402](https://agentdrives.com/standards/x402) (pay per request)
- Steps: [Find](https://agentdrives.com/access/find), [Get in](https://agentdrives.com/access/get-in), [Use](https://agentdrives.com/access/use), [Pay](https://agentdrives.com/access/pay) (first stop), [Leave](https://agentdrives.com/access/leave)
- Topics: [Docs and llms.txt](https://agentdrives.com/access/find/docs) (Find step), [Sign-up](https://agentdrives.com/access/get-in/sign-up) (Get in step), [MCP and API](https://agentdrives.com/access/use/mcp-and-api) (Use step), [Card checkout](https://agentdrives.com/access/pay/card-checkout) (Pay step), [Delete the account](https://agentdrives.com/access/leave/delete-account) (Leave step)
- Rivals: [Checkout.com](https://agentdrives.com/tools/checkout?via=md) (L3), [GoCardless](https://agentdrives.com/tools/gocardless?via=md) (L3), [Nevermined](https://agentdrives.com/tools/nevermined?via=md) (L3)

The level follows [these rules](https://agentdrives.com/changes#rules-2026-11-18).

## Level by drive

One drive so far, on 6 Nov 2026.

[Every drive in full](https://agentdrives.com/tools/crossmint/history?via=md)

Source: our drives of Crossmint. As of 6 Nov 2026.

One drive so far.

Data: [JSON](https://agentdrives.com/api/tools/crossmint.json?via=md) · [Markdown](https://agentdrives.com/tools/crossmint.md?via=md) · [History](https://agentdrives.com/tools/crossmint/history?via=md) · [Atom](https://agentdrives.com/feeds/tools/crossmint.atom) · [JSON Feed](https://agentdrives.com/feeds/tools/crossmint.json) · licence [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) · evidence sha256 `969a4c5a4332584a54b2f8dd6fd4315e0dc29d4bcdf5346f30d19795a55bbe00`

**Found an error?** [Tell us on this page](https://agentdrives.com/tools/crossmint?via=md#correct) or read [how corrections work](https://agentdrives.com/corrections?via=md).
