# Can an AI agent use FusionAuth? L4 · Runs on its own

**L4 · Runs on its own** · [What the levels mean](https://agentdrives.com/method/levels?via=md)

Rank among the 246 products we drove:

- Holds up: 92 / 100, tied #15 (Reliability under errors)
- Steps on its own: 4 / 5, tied #3 (Find to Leave, no person)
- Core work done: 100%, tied #1 (7 of 7 core tasks)
- Time to get in: 9 minutes, tied #93 (Getting in on its own)

**Can an AI agent use FusionAuth? Yes.** Our agent got into FusionAuth on its own without signing up and used its API and MCP server. No CLI found.

- Level confirmed: Confirmed (The drives of 5 Nov and 15 Nov 2026 agree)
- Got in by: Runs locally (9 minutes, no account)
- Got as far as: Use: all 7 core tasks (No payment made)

Category: Auth and identity · Connects by: MCP, API, SDK · Agent payments: None found · Driven: 15 Nov 2026

**How to connect**
- MCP server URL: `https://ai.fusionauth.dev/mcp/docs` (documented). Source: https://ai.fusionauth.dev/mcp/docs
- API base URL: `https://sandbox.fusionauth.io` (documented). Source: https://fusionauth.io/docs/openapi.yaml
- CLI install: `npm install -g @fusionauth/cli` (documented). Source: https://www.npmjs.com/package/@fusionauth/cli
- Docs page: `https://fusionauth.io/docs/` (tested in our drive). Source: https://fusionauth.io/docs/

Checked 18 Nov 2026: MCP server answers, 2 tools; sign-up page, llms.txt and OpenAPI unchanged since the check before.

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

Got in on 2 of the last 3 drives. Enough for Level 4: the newest drive and half or more of the last three got in.

## 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, an OpenAPI and an MCP server. (453 doc pages read)
2. **Get in**: On its own. Got in without signing up. Email and password: stopped by a CAPTCHA. (9 minutes)
3. **Use**: On its own. Did: create tenants, applications, roles and OAuth clients; register users and sign them in. (7 of 7 core tasks)
4. **Pay**: Not tried. Paying needs an account.
5. **Leave**: On its own. Nothing to cancel. It deletes what it made.

To reach L5: Pay without a person.

## How an agent connects

MCP, API, OpenAPI, SDK and llms.txt found.

- **MCP server**: 249 of 310 tested. Official server. @fusionauth/mcp-api 0.1.14 (serverInfo 0.1.13), stdio, 310 tools when driven, the 25 Nov 2025 MCP spec; also a docs MCP io.fusionauth/mcp-docs. (310 tools mapped in our drive) 2 tools listed without signing in on 30 Oct 2026: `ask_fusionauth_question`, `search_fusionauth_sources`.
- **API**: 287 of 364 tested. REST, 364 operations, API key auth. (364 operations mapped in our drive)
- **OpenAPI**: Yes. https://fusionauth.io/docs/openapi.yaml, openapi 3.0.3, info.version 1.70.0.
- **CLI**: None found.
- **SDK**: All tested. Packages: npm [@fusionauth/typescript-client](https://www.npmjs.com/package/@fusionauth/typescript-client), npm [@fusionauth/angular-sdk](https://www.npmjs.com/package/@fusionauth/angular-sdk). (1 method mapped in our drive)
- **llms.txt**: Yes. https://fusionauth.io/llms.txt
- **Agent payments**: None found. No payment made: paying needs an account.
- **Docs**: In part. Covers getting started, auth, error codes and code examples. Describes each operation only in part: completeness 78 of 100. 7 of 7 documented ways to do the core work worked. https://fusionauth.io/docs/ (453 of 849 pages read)

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

## Getting in

Our agent tried 2 of 4 ways in; 1 worked without a person.

The way that worked: **Runs locally**, 9 minutes. No inbox, browser or person needed.

- Agent on a server (API, MCP, CLI): Partly. Signs up, gets an API key, makes its first API call and does all of the core work. Payment wasn't tried.
- Assistant in its own browser (Like ChatGPT agent): Partly. Signs up and does all of the core work. Payment wasn't tried.
- Signed agent: Not checked: the robots.txt on account.fusionauth.io asks crawlers to stay off the sign-up page.

| Way in | What happened |
|---|---|
| Runs locally | Worked |
| Email and password | Stopped by a CAPTCHA |
| Added by an admin | Not tried: Needs a person's account |
| A key from a person | Not tried: Needs a person's account |

- Crawlers: **The robots.txt on account.fusionauth.io asks crawlers to stay off sign-up.** Nothing to fix.

[How agents get past bot checks](https://agentdrives.com/guides/let-signed-agents-past-your-bot-check.md?via=md)

## What an agent can do with FusionAuth

All 7 core tasks done, and 9 of 10 others.

- Holds up: 92 of 100 (Reliability 88, Handles mistakes 96). The mean of the two.
- Docs match: 93 of 100 (Docs accuracy 100, Docs completeness 78). Docs accuracy counts 70%, completeness 30%.
- Surface tested: 68 of 100. 501 of 734 passed: the 751 documented operations less the 17 that spend money or reach real people

Core tasks:

| Task | Result | Done through |
|---|---|---|
| Register and sign in an end user | Done | API, MCP |
| Configure a sign-in method (social, passkey, passwordless, LDAP) | Done | API |
| Configure SAML SSO for an application | Done | API |
| Assign a role or permission to a user or group | Done | API |
| Register an OAuth application/client | Done | API |
| Create a tenant to isolate a set of users and applications | Done | API, MCP |
| Enable multi-factor authentication for a user | Done | API |

9 of 10 other tasks done:

- Create a user group and add members: Done
- Register a webhook for authentication lifecycle events: Done
- Read the audit log: Done
- Create an email template for auth emails: Done
- Import or export users in bulk: Done
- Add custom logic with a Lambda hook: Done
- Restrict API/admin access by IP address: Done
- Set an application login/password security policy: Done
- Set custom branding/theme on hosted login pages: Done
- Provision users via SCIM: Paid plan
- Connect a legacy user database or LDAP system: Paid plan
- Register a machine-to-machine entity for client-credentials auth: Paid plan
- Score a login's risk and trigger step-up MFA: Paid plan
- Check a user's identity document (identity verification): Blocked. FusionAuth offers email and phone verification, not document checks.

The plans they need:

- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license
- : FusionAuth paid license

The drive used the free plan (self-hosted Community edition, no license key).

What it can't do, and why:

- Sign up for FusionAuth Cloud or buy a license: The account portal has a visible reCAPTCHA and there is no agent payment route.
- Use SCIM, machine-to-machine entities, LDAP connectors or risk-based MFA: These need a paid license; the free edition refuses them.
- Check a user's identity document: FusionAuth offers email and phone verification, not document checks.

**Surface audit:** 751 documented operations: 581 run, 153 blocked by sandbox or plan limits and 17 left out because they spend money or reach real people. Of the 581 run, 501 passed and 80 didn't; 2 behaved differently from the docs. By surface, 287 on the API, 249 on the MCP server, 1 in the SDK and 44 elsewhere, such as sign-in and webhooks.

## Tips for agent builders

- Use the self-hosted Community edition: docker images fusionauth/fusionauth-app plus Postgres, and a Kickstart file whose apiKeys entry sets keyManager: true so you can mint scoped keys
- Set SEARCH_TYPE=database to run without Elasticsearch
- Send X-FusionAuth-TenantId only on tenant-scoped calls once there is more than one tenant; global objects refuse it
- Add ?hardDelete=true when deleting applications and user actions, else they are only deactivated
- Treat 400 [notLicensed] as a plan gate, not a bad request
- For the MCP server, wrap user links in identityProviderLink, pass query booleans as strings, and use REST for group members, messengers, connectors and JWT-auth routes

## Paying

Ways to pay checked 15 Nov 2026.

- Result: Not tried. Paying needs an account.

## Leaving

- Result: On its own. Nothing to cancel. It deletes what it made.
- Its account: Not applicable: our agent used FusionAuth without an account, and it deletes what it made.

## Checked every day

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

- Sign-up page, loads on the newest check; 14 of 14 days
- llms.txt, found on the newest check; 14 of 14 days
- MCP server, answers on the newest check (2 tools); 14 of 14 days
- OpenAPI, found on the newest check; 14 of 14 days

## How FusionAuth compares

In Auth and identity: Ranks 1 of the 1 ranked in Auth and identity, by confirmed level. Not ranked: 1 our agent couldn't get into on its own; 7 await a second drive.

- **FusionAuth** (this page): L4, Gets in on its own
- [PropelAuth](https://agentdrives.com/tools/propelauth.md?via=md): L1, Can't sign up on its own
- [SuperTokens](https://agentdrives.com/tools/supertokens.md?via=md): L4, Gets in on its own
- [Clerk](https://agentdrives.com/tools/clerk.md?via=md): L3, Gets in on its own
- [WorkOS](https://agentdrives.com/tools/workos.md?via=md): L3, Gets in on its own
- [Auth0](https://agentdrives.com/tools/auth0.md?via=md): L3, Gets in on its own
- [Kinde](https://agentdrives.com/tools/kinde.md?via=md): L3, Gets in on its own
- [HumanMirror](https://agentdrives.com/tools/humanmirror.md?via=md): L3, Gets in on its own
- [Web3Auth (MetaMask Embedded Wallets)](https://agentdrives.com/tools/web3auth.md?via=md): L3, Gets in on its own

[All 9 in Auth and identity](https://agentdrives.com/categories/auth?via=md)

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

On our standard (28 criteria). Best or ahead on 4 of 28 criteria, behind on 6.

- [Delete by API, checked](https://agentdrives.com/standard#leave-02): **The best result** All 7 others (auth and identity products) lower.
- [Typed, tight input schemas](https://agentdrives.com/standard#mcp-06): **Ahead of 77 of 91** others (every product), 14 the same.
- [A tool list that fits](https://agentdrives.com/standard#mcp-11): **Ahead of 48 of 91** others (every product), 43 the same.
- [The best way in is built for agents](https://agentdrives.com/standard#in-12): **Level with 4 of 7** others (auth and identity products), 3 lower.
- [Sign-in for agents acting for a person](https://agentdrives.com/standard#in-08): **Level with 3 of 5** others (auth and identity products), 2 lower.
- [Results sized for a context window](https://agentdrives.com/standard#mcp-09): **Ahead of 3 of 5** others (auth and identity products), 1 the same, 1 higher.
- [No person in the buy path](https://agentdrives.com/standard#pay-06): **Level with 4 of 5** others (auth and identity products), 1 lower.
- [A way in with no person](https://agentdrives.com/standard#in-01): **Level with 7 of 8** others (auth and identity products), 1 lower.
- [Tool names that sort themselves](https://agentdrives.com/standard#mcp-03): **Level with 85 of 91** others (every product), 6 lower.
- [llms.txt that works](https://agentdrives.com/standard#find-02): **Level with 6 of 8** others (auth and identity products), 1 lower, 1 higher.
- [Terms allow an agent's account](https://agentdrives.com/standard#in-10): **The same as all** 8 others (auth and identity products).
- [Structured results](https://agentdrives.com/standard#mcp-08): **In the middle** 2 of 5 others (auth and identity products) lower, 1 the same, 2 higher.
- [Export before leaving](https://agentdrives.com/standard#leave-03): **In the middle** 2 of 7 others (auth and identity products) lower, 3 the same, 2 higher.
- [WebMCP tools on the web app](https://agentdrives.com/standard#proto-02): **The same as all** 5 others (auth and identity products).
- [An agent payment rail on the core](https://agentdrives.com/standard#pay-02): **Level with 7 of 8** others (auth and identity products), 1 higher.
- [Pages fit one fetch](https://agentdrives.com/standard#docs-08): **Level with 6 of 7** others (auth and identity products), 1 higher.
- [Spend an agent can see and cap](https://agentdrives.com/standard#pay-05): **Level with 6 of 7** others (auth and identity products), 1 higher.
- [A plan an agent can buy by API](https://agentdrives.com/standard#pay-03): **Level with 5 of 6** others (auth and identity products), 1 higher.
- [Agents pick the right tool](https://agentdrives.com/standard#mcp-05): **Level with 72 of 91** others (every product), 19 higher.
- [Docs an agent can navigate](https://agentdrives.com/standard#docs-11): **In the middle** 1 of 7 others (auth and identity products) lower, 3 the same, 3 higher.
- [Every way in works for an agent](https://agentdrives.com/standard#in-11): **Level with 5 of 7** others (auth and identity products), 2 higher.
- [Agents may read docs, API and sign-up](https://agentdrives.com/standard#find-01): **Level with 3 of 5** others (auth and identity products), 2 higher.
- [One way to page through lists](https://agentdrives.com/standard#api-03): **Tied for the lowest result** 2 of 4 others (auth and identity products) the same, 2 higher.
- [A path written for agents](https://agentdrives.com/standard#docs-10): **Behind 4 of 7** others (auth and identity products), 3 the same.
- [Versions an agent can pin](https://agentdrives.com/standard#api-07): **Behind 3 of 5** others (auth and identity products), 2 the same.
- [Truthful annotations](https://agentdrives.com/standard#mcp-07): **Behind 68 of 91** others (every product), 23 the same.
- [Descriptions that say when to use the tool](https://agentdrives.com/standard#mcp-04): **Behind 70 of 91** others (every product), 21 the same.
- [Prices an agent can read](https://agentdrives.com/standard#find-06): **Behind 6 of 7** others (auth and identity products), 1 the same.

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

## Our agent didn't try Pay

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

Not every fix clears a blocker: some make a step that works easier for agents.

Agent blockers: **5 found, none stopped the drive** Agent surfaces: **2 issues found**

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

We mail a private link to an address at fusionauth.io.

1. Find Clear
2. Get in Clear
3. Use Clear
4. Pay Not tried
5. Leave Clear

Or [claim FusionAuth](https://agentdrives.com/go/claim/fusionauth?add=fusionauth.io&ref=listing) to see where each issue is and ask for a free drive after a fix.

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

### On our standard

Best or ahead on 4 of 28 criteria, behind on 6. Each against the products we measured on it.

- [Delete by API, checked](https://agentdrives.com/standard#leave-02)
- [Typed, tight input schemas](https://agentdrives.com/standard#mcp-06)
- [A tool list that fits](https://agentdrives.com/standard#mcp-11)
- [The best way in is built for agents](https://agentdrives.com/standard#in-12)
- [Versions an agent can pin](https://agentdrives.com/standard#api-07)
- [Truthful annotations](https://agentdrives.com/standard#mcp-07)
- [Descriptions that say when to use the tool](https://agentdrives.com/standard#mcp-04)
- [Prices an agent can read](https://agentdrives.com/standard#find-06)

**All 28 criteria**

- [Sign-in for agents acting for a person](https://agentdrives.com/standard#in-08)
- [Results sized for a context window](https://agentdrives.com/standard#mcp-09)
- [No person in the buy path](https://agentdrives.com/standard#pay-06)
- [A way in with no person](https://agentdrives.com/standard#in-01)
- [Tool names that sort themselves](https://agentdrives.com/standard#mcp-03)
- [llms.txt that works](https://agentdrives.com/standard#find-02)
- [Terms allow an agent's account](https://agentdrives.com/standard#in-10)
- [Structured results](https://agentdrives.com/standard#mcp-08)
- [Export before leaving](https://agentdrives.com/standard#leave-03)
- [WebMCP tools on the web app](https://agentdrives.com/standard#proto-02)
- [An agent payment rail on the core](https://agentdrives.com/standard#pay-02)
- [Pages fit one fetch](https://agentdrives.com/standard#docs-08)
- [Spend an agent can see and cap](https://agentdrives.com/standard#pay-05)
- [A plan an agent can buy by API](https://agentdrives.com/standard#pay-03)
- [Agents pick the right tool](https://agentdrives.com/standard#mcp-05)
- [Docs an agent can navigate](https://agentdrives.com/standard#docs-11)
- [Every way in works for an agent](https://agentdrives.com/standard#in-11)
- [Agents may read docs, API and sign-up](https://agentdrives.com/standard#find-01)
- [One way to page through lists](https://agentdrives.com/standard#api-03)
- [A path written for agents](https://agentdrives.com/standard#docs-10)

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

## For the team behind FusionAuth

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

Not every fix clears a blocker: some make a step that works easier for agents. Agent blockers: 5 found, none stopped the drive · Agent surfaces: 2 issues found.

Our agent didn't try Pay. Its team gets the two findings that matter most in full, free, by a private link to an address at fusionauth.io.

- Find: Clear
- Get in: Clear
- Use: Clear
- Pay: Not tried
- Leave: Clear

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

## Related research

Guides and findings on the Find step.

- [Content Signals and AI rules in SaaS robots.txt](https://agentdrives.com/findings/content-signals?via=md) (finding): 57 of the 245 products whose robots.txt our drives read (23%) tell AI there what it may do with their content, all with Cloudflare's Content-Signal line.
- [llms.txt, OpenAPI and MCP alone don't let AI agents in](https://agentdrives.com/guides/agent-ready-files-dont-open-the-door?via=md) (guide): 212 of 246 products publish an llms.txt and 131 of 246 an OpenAPI file; our agent still didn't get into 137 of 246 on its own.
- [Make your SaaS agent-ready: the whole path, step by step](https://agentdrives.com/guides/make-your-saas-agent-ready?via=md) (guide): Docs, sign-up, the key, MCP, paying and leaving: what an agent needs at each step, with the fix for each.

[Everything on Find](https://agentdrives.com/access/find)

## Articles that cite FusionAuth

- [How to let AI agents sign up for your product](https://agentdrives.com/guides/let-ai-agents-sign-up?via=md)

## Questions people ask about FusionAuth

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

Yes, over the API, without a browser. Without a browser on the local way: the official Docker images plus a documented Kickstart file created the admin and a superuser API key on first boot.

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

Yes. FusionAuth has an official MCP server. A new agent can get in on its own: in the drive of 15 Nov 2026, our agent signed up and got its key without a person. Through the MCP server, our agent tried 2 tasks and did 2.

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

In the drive of 15 Nov 2026, our agent got its FusionAuth API key over the API, without a browser or a person.

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

Not applicable: our agent used FusionAuth without an account, and it deletes what it made.

## Where FusionAuth fits

- Category: [Developer platform](https://agentdrives.com/categories/developer?via=md), [Auth and identity](https://agentdrives.com/categories/auth?via=md)
- Tasks: [Register a webhook](https://agentdrives.com/tasks/register-a-webhook), [Set a member's role or permissions](https://agentdrives.com/tasks/set-a-member-s-role-or-permissions), [Import data from a file or another tool](https://agentdrives.com/tasks/import-data-from-a-file-or-another-tool), [Read the audit log](https://agentdrives.com/tasks/read-the-audit-log), [Register an OAuth app](https://agentdrives.com/tasks/register-an-oauth-app), [Set the brand logo and colours](https://agentdrives.com/tasks/set-the-brand-logo-and-colours)
- Wall: [Bot checks](https://agentdrives.com/access/get-in/bot-checks) (our agent got in another way), [Let signed AI agents past your bot check](https://agentdrives.com/guides/let-signed-agents-past-your-bot-check?via=md) (fix guide)
- Vendors: [Google reCAPTCHA](https://agentdrives.com/research/bot-checks?via=md#recaptcha) (on one way in)
- Standards: [MCP](https://agentdrives.com/standards/mcp) (tools an agent can call), [OpenAPI](https://agentdrives.com/standards/openapi) (its API, described), [llms.txt](https://agentdrives.com/standards/llms-txt) (its docs, listed for agents)
- 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), [Leave](https://agentdrives.com/access/leave)
- Topics: [MCP servers](https://agentdrives.com/access/find/mcp-servers) (Find step), [Docs and llms.txt](https://agentdrives.com/access/find/docs) (Find step), [MCP sign-in](https://agentdrives.com/access/get-in/mcp-sign-in) (Get in step), [Sign-up](https://agentdrives.com/access/get-in/sign-up) (Get in step)
- Rivals: [SuperTokens](https://agentdrives.com/tools/supertokens?via=md) (L4), [Auth0](https://agentdrives.com/tools/auth0?via=md) (L3), [Clerk](https://agentdrives.com/tools/clerk?via=md) (L3)

## Level by drive

3 drives since 1 Nov 2026.

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

Source: our drives of FusionAuth. As of 15 Nov 2026.

L4 in both drives since 5 Nov 2026.

## Cited in 1 article

- [How to let AI agents sign up for your product](https://agentdrives.com/guides/let-ai-agents-sign-up?via=md)

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

Data: [JSON](https://agentdrives.com/api/tools/fusionauth.json?via=md) · [Markdown](https://agentdrives.com/tools/fusionauth.md?via=md) · [History](https://agentdrives.com/tools/fusionauth/history?via=md) · [Atom](https://agentdrives.com/feeds/tools/fusionauth.atom) · [JSON Feed](https://agentdrives.com/feeds/tools/fusionauth.json) · licence [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) · evidence sha256 `2c314393e79c654d4f41ab9e77601e269ee377680ac2dcfe7f7d4d5da55ba332`

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