No tracking. No cookie wall.·100 % EU-hosted on Hetzner
Use casesSecurity

STRIDE threat model — quick narrative via search

You are designing a new B2B authentication flow using OAuth 2.1 + OIDC + Keycloak. Your security architect wants a fast STRIDE narrative grounded in OWASP and STRIDE-patterns before committing to a full workflow run.

Citations8
Tool calls4
Captured withclaude-opus-4-7

How the gateway answered

Captured sessionclaude-opus-4-7 · team tier · gateway.ansvar.eu
You

Build a STRIDE threat model for the following authentication architecture: user → Keycloak → resource server → PostgreSQL. List threats per component, mitigations, and cite OWASP / NIST 800-53 / STRIDE patterns where relevant.

Ansvar gateway — 4 tool calls
search(query="OAuth authentication token theft replay spoofing" sources=[OWASP MCP] limit=6)6 results from OWASP MCP — Session Timeout (×4), Weak Encryption (×2)
search(query="identity authentication access control NIST 800-53 IA…" sources=[Security Controls MCP] limit=6)1 weakly-relevant result from Security Controls MCP; insufficient for NIST 800-53 citations
search(query="PostgreSQL database SQL injection privilege escalation" sources=[OWASP MCP] limit=4)4 results from OWASP MCP — Testing for SQL Injection + Objectives 1/2/3
search(query="Keycloak OIDC session hijacking refresh token" sources=[OWASP MCP, STRIDE MCP] limit=5)10 results: OWASP Session Fixation/Logout/Management/Hijacking + STRIDE OAuth replay, Predictable Tokens, WebSocket Origin Bypass
Cited answer

STRIDE threat model — OAuth 2.1 + OIDC + Keycloak

Four components in scope: user agent → Keycloak (auth server) → resource server → PostgreSQL. STRIDE per component with mitigations grounded in the gateway's OWASP WSTG corpus and the STRIDE MCP threat-pattern library.

1. User agent ↔ Keycloak (login + token issuance)

CategoryThreatCited mitigation
SpoofingStolen credentials replayed at loginOAuth 2.1 PKCE; MFA. See OWASP — Testing Session Fixation
SpoofingOIDC redirect-URI tampering steers tokens to an attackerStrict redirect-URI allowlist; exact-match only. See STRIDE MCP — OAuth 2.0 Access Token Theft and Replay
TamperingAuth-code interception in the redirectPKCE S256 challenge binds the code to the originating client. Same OWASP/STRIDE references.
RepudiationUser denies they consented to the scopes they were grantedKeycloak's consent screen + audit log of grants. (No on-point citation in this gateway run — flagged for review.)
Information disclosureRefresh token leak grants long-lived accessSTRIDE MCP — Predictable Token Generation via Weak Random Number Generators. Use rotating refresh tokens; bind tokens to the client …

Every claim traces to a source you can open

A typical AI assistant invents a citation that looks plausible. Ansvar retrieves the real one. Every finding above rests on one of these 8 primary sources — official legislation, agency guidance, and case law — each a link you or your auditor can open and check. Nothing here is generated.

Run this on your own data

This is available as a service: Threat Model as a Service. Bring your own documents and scope, and we'll run it end-to-end — every finding cited and validated by the expert who delivers it.