Stytch MCP
Stytch MCP lets authorized agents configure Stytch projects, redirect URLs, SDK settings, email templates, API keys, and other workspace-management resources.
Verdict
Common use cases
- Investigate failed login attempts during support calls
- Audit user sessions for security reviews
- Check authentication status before account changes
- Retrieve user profile data for onboarding issues
- Monitor active sessions across devices
Integration
- Vendor
- Stytch MCP
- Category
- other
- Auth
- OAUTH2
- Composio slug
stytch_mcp
Tools
Setup
Setup guide
- 11. Open your Switchy workspace settings and navigate to the Integrations tab. 2. Search for Stytch MCP and click Connect. 3. You'll be redirected to Stytch's OAuth consent screen — sign in with an account that has read access to your Stytch project. 4. Grant the requested scopes, which typically include reading user data, sessions, and authentication events. 5. After authorizing, you'll return to Switchy and see Stytch MCP listed as Connected. 6. Open any Project and type '@Stytch' followed by a question like 'show me active sessions for user ID abc123' to confirm the connection works. 7. If the MCP responds with session data, setup is complete and your team can start querying user authentication details directly in conversations.
What teammates see: by default, memories from Stytch MCP are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Check User Sessions
@Stytch show me all active sessions for user ID [user_id]. Include device type, IP address, and last activity timestamp.
Audit Recent Logins
@Stytch list all authentication events in the last 24 hours. Group by success and failure, and highlight any unusual IP addresses.
Verify User Profile
@Stytch retrieve the full profile for user email [email@example.com]. Show email verification status, phone number, and any custom metadata.
Investigate Failed Auth
@Stytch find all failed authentication attempts for user ID [user_id] in the past week. Include error codes and timestamps.
Monitor Session Activity
@Stytch show session creation and expiration events for the past 48 hours. Flag any sessions created from new geographic regions.
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Stytch analyze the last 50 authentication events and summarize any patterns that might indicate credential stuffing or account takeover attempts
Reviewed 50 recent auth events (Jan 10-15, 2024). Found 3 anomalies: 12 failed login attempts from IP 203.0.113.45 targeting 8 different user accounts within a 2-minute window on Jan 12, suggesting automated credential testing. All attempts used email/password flow. Recommend enabling rate limiting on this IP range and reviewing those 8 accounts for compromised credentials. No successful breaches detected.
Pairs Stytch's event logs with AI reasoning to surface security threats. The MCP provides raw event data; the AI performs pattern detection. This workflow requires read access to authentication logs and works best when Stytch's webhook events are enabled for real-time monitoring.
Use-case deep-dives
Stytch MCP during live auth degradation incidents
A 4-person engineering team gets paged at 2am: users report they can't log in. The on-call engineer needs to check if Stytch is rejecting requests, if rate limits are hit, or if a recent API key rotation broke the integration. Stytch MCP gives the engineer a conversational interface to query error rates, inspect recent API calls, and compare current traffic to baseline—all from Switchy's incident channel instead of context-switching to the Stytch dashboard. The limitation: if your auth stack spans multiple providers (Stytch for passwordless, Okta for enterprise SSO), the MCP only covers the Stytch side. For teams running Stytch as their sole auth layer and triaging fewer than 5 incidents a month, it shaves 10 minutes off mean-time-to-diagnosis.
Frequently asked
What does the Stytch MCP do in Switchy?
The Stytch MCP connects your team's Stytch authentication data to Switchy's AI workspace. It lets AI assistants query user sessions, authentication events, and identity data without switching contexts. Since Stytch handles passwordless auth and magic links, this MCP is most useful for teams debugging login flows or auditing user access patterns across their product.
Does Stytch MCP require admin OAuth permissions?
Yes. Stytch uses OAuth2, and the MCP needs read access to your project's users, sessions, and authentication logs. You'll authorise through Stytch's standard OAuth flow, which requires project-level permissions. If you're not an admin in your Stytch dashboard, you won't be able to complete the connection.
Can the Stytch MCP create or delete users?
No. The MCP is read-only by design. It can surface user records, session states, and auth event logs, but it won't modify your Stytch project. If you need to provision users or revoke sessions, use Stytch's dashboard or API directly. This keeps your production auth layer safe from accidental AI actions.
Why use this instead of Stytch's API directly?
The MCP saves you from writing custom scripts every time someone asks 'how many users logged in this week' or 'which sessions are still active'. Your team can ask those questions in plain English inside Switchy, and the AI handles the API calls. It's faster for ad-hoc queries, slower for automated reporting pipelines.
Who on the team should connect the Stytch MCP?
Whoever owns your Stytch project and understands which user data is safe to expose in a shared workspace. Usually that's your engineering lead or security engineer. Once connected, anyone in your Switchy workspace can query the data, so treat this like granting dashboard access to the whole team.