How is this different from connecting Stytch MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Stytch MCP API changes?

We track the Stytch MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Stytch tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Stytch?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Stytch MCP server myself?

No. TrustGate connects to the Stytch MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Stytch directly.

Connect Stytch MCP to your AI agents, securely.

Q: Is Stytch MCP secure?

The server itself exposes 21 tools, and some of them can write to your Stytch account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Stytch.

One secured endpoint for every Stytch call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Stytch MCP now

What your agents can ask once Stytch MCP is connected.

Select the tools your AI agents can call

21 tools from the Stytch MCP schema.

Showing 21 tools

getAllRedirectURLsreadLists the redirect URLs configured on the project, meaning the destinations a user can be sent to after an authentication flow completes.
createRedirectURLhigh riskAdds a redirect URL to the project's allowlist. Since redirect URLs are the destinations an authentication token may be sent to after login, adding one is the standard route to account takeover.
updateRedirectURLhigh riskUpdates an existing redirect URL, changing where a completed authentication flow sends the user.
deleteRedirectURLhigh riskRemoves a redirect URL from the project, which breaks any login flow that relies on it.
getAllPublicTokensreadLists the project's public tokens, which identify the project to client-side code.
createPublicTokenwriteCreates a public token for the project. Public tokens are client-side identifiers rather than secrets, which makes this the mildest write on the connection.
deletePublicTokenhigh riskDeletes a public token, breaking whatever frontend is currently authenticating with it.
createSecrethigh riskCreates a project secret, which is the server-side credential a backend uses to authenticate against Stytch.
deleteSecrethigh riskDeletes a project secret, which can take production sign-in down immediately if the secret is in use.
getAllEmailTemplatesreadLists the email templates configured on the project.
createEmailTemplatehigh riskCreates an email template, which defines the content of authentication messages such as magic links and password resets that real end users receive.
updateEmailTemplatehigh riskUpdates an email template, changing the wording, branding or links in a security-critical message users are primed to click.
deleteEmailTemplatehigh riskDeletes an email template, which can fall back to a default or break the message flow that depends on it.
getConsumerSDKConfigreadReturns the Consumer SDK configuration, which governs what the client SDK is permitted to do against the project.
getB2BSDKConfigreadReturns the B2B SDK configuration for organization-based authentication.
updateConsumerSDKConfighigh riskUpdates the Consumer SDK configuration, changing which authentication methods a browser or app may use against the live product.
updateB2BSDKConfighigh riskUpdates the B2B SDK configuration, changing the authentication posture for organization-based sign-in.
listProjectsreadLists the projects in the Stytch workspace.
getPasswordStrengthConfigreadReturns the password strength configuration governing what passwords the project will accept.
createProjecthigh riskCreates a new Stytch project, standing up a new authentication environment in the workspace.
setPasswordStrengthConfighigh riskSets the password strength configuration, which changes the floor for every password set after it. A weakening is invisible to users who already have an account.

What Stytch MCP exposes

Stytch MCP groups its 21 tools into 5 categories, each with a suggested starting policy.

All 21 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Credentials	Creating, listing and deleting public tokens, which identify a project to client-side code, and creating or deleting project secrets, which are the server-side credential a backend authenticates with	Read / Write	createSecret mints a credential and deleteSecret revokes one, either of which can take production sign-in down or leave a key in circulation. Deleting a public token breaks whatever frontend is currently using it. Gate all three and alert on them
Redirect URLs	Creating, listing, updating and deleting the redirect URLs a project will send users to after a magic link, OAuth flow or password reset completes	Read / Write	This is the allowlist of destinations an authentication token may be sent to. Adding an attacker-controlled URL is the standard account-takeover path, and removing a legitimate one breaks login for real users. Treat every write here as a security change requiring named approval
Email Templates	Creating, listing, updating and deleting the email templates the project sends, which cover magic links, password resets and other authentication messages	Read / Write	These are the messages end users receive and act on, in a context where they are already primed to click a link. Changing the wording, branding or links in one is closer to editing a phishing surface than to editing marketing copy
SDK Configuration	Reading and updating the Consumer SDK and B2B SDK configuration, which governs which authentication methods the client SDK may use and the conditions under which it operates	Read / Write	SDK configuration decides what a browser or app is permitted to do against the project. Read both before changing either, and treat an update as a change to the authentication posture of the live product rather than as a settings tweak
Projects & Password Policy	Creating a project and listing the projects in the workspace, plus reading and setting the password strength configuration that governs what passwords the project will accept	Read / Write	setPasswordStrengthConfig changes the floor for every password set after it, and a weakening is invisible to existing users. createProject stands up a new authentication environment, which is cheap to call and easy to forget about

Connect Stytch MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Stytch
Find Stytch in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Stytch MCP or block it. You allow the exact tools each agent can call, one by one.

Stytch MCP	21 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 21 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Stytch MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Stytch MCP21 tools at mcp.stytch.dev/mcp, sign-in required. The server manages the configuration of a Stytch Project, which is the authentication layer of the organization's own product, so almost every write changes how that product's end users sign in. No read-only mode, scope split or per-tool gating is published for this server, which leaves the narrowing to the gateway in front of it.
Coding assistants	Connection & credentialsper user	Credentials5 tools
Internal copilots	Tool access policiesdeny by default	Redirect URLs4 tools
Workflow automations	Request & response inspectioninline	Email Templates4 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools8 tools