/

CVE-2026-63094

Severity:

HIGH RISK

(8.1)

Open Redirect (CWE-601) in SigNoz SSO OAuth Flow

CVE-2026-63094

Deferred

Vulnerability type

Category:

Open Redirect / Authentication Flow Abuse

CWE:

CWE-345, CWE-601

Impact:

Account Takeover via Token Theft

Severity:

HIGH RISK

(8.1)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

SigNoz contains an open redirect flaw in its SSO authentication flow that lets attackers craft login URLs which send users to attacker-controlled domains after SSO completes.

On instances using Google OAuth, SAML, or OIDC, a malicious redirect can expose access and refresh tokens to unauthenticated attackers who control the destination host.

SigNoz contains an open redirect flaw in its SSO authentication flow that lets attackers craft login URLs which send users to attacker-controlled domains after SSO completes.

On instances using Google OAuth, SAML, or OIDC, a malicious redirect can expose access and refresh tokens to unauthenticated attackers who control the destination host.

Affected context

This vulnerability appears in SigNoz deployments that:

  • Enable SSO using Google OAuth, SAML, or OIDC providers
  • Expose an unauthenticated sessions context endpoint used to construct login URLs
  • Accept a ref or redirect parameter from client input without strict validation
  • Embed access and refresh tokens in redirect flows that can be sent to arbitrary domains

It becomes relevant whenever users follow SSO login links generated by or relayed through untrusted parties.

This vulnerability appears in SigNoz deployments that:

  • Enable SSO using Google OAuth, SAML, or OIDC providers
  • Expose an unauthenticated sessions context endpoint used to construct login URLs
  • Accept a ref or redirect parameter from client input without strict validation
  • Embed access and refresh tokens in redirect flows that can be sent to arbitrary domains

It becomes relevant whenever users follow SSO login links generated by or relayed through untrusted parties.

Exploit preconditions

  • Attack surface: Public SigNoz web interface and SSO login endpoints
  • Required access: None; attacker can be fully unauthenticated
  • Trigger point: Victim clicking a crafted SigNoz SSO login URL that includes a malicious ref redirect parameter
  • Why exploitation is possible: The sessions context endpoint trusts the caller-provided ref value and incorporates it into the post-authentication redirect, allowing tokens or session context to be sent to attacker-controlled hosts rather than constrained, known application origins.
  • Attack surface: Public SigNoz web interface and SSO login endpoints
  • Required access: None; attacker can be fully unauthenticated
  • Trigger point: Victim clicking a crafted SigNoz SSO login URL that includes a malicious ref redirect parameter
  • Why exploitation is possible: The sessions context endpoint trusts the caller-provided ref value and incorporates it into the post-authentication redirect, allowing tokens or session context to be sent to attacker-controlled hosts rather than constrained, known application origins.

Why this happens

The issue stems from insufficient validation of redirect targets in the SSO login flow.

The sessions context endpoint accepts a ref parameter and uses it in constructing a login URL without enforcing a strict allowlist of trusted domains or paths.

This aligns with CWE-601: Open Redirect and CWE-345: Insufficient Verification of Data Authenticity, where untrusted, client-controlled data is used to determine navigation and token delivery behavior, letting attackers interpose their own endpoints into sensitive authentication flows.

The issue stems from insufficient validation of redirect targets in the SSO login flow.

The sessions context endpoint accepts a ref parameter and uses it in constructing a login URL without enforcing a strict allowlist of trusted domains or paths.

This aligns with CWE-601: Open Redirect and CWE-345: Insufficient Verification of Data Authenticity, where untrusted, client-controlled data is used to determine navigation and token delivery behavior, letting attackers interpose their own endpoints into sensitive authentication flows.

Potential impact

  • Attackers may steal access and refresh tokens for arbitrary SigNoz users.
  • Compromised tokens could allow full account takeover for impacted users.
  • Attackers can access or modify observability data and dashboards associated with victim accounts.
  • Malicious actors might pivot into other integrated systems that trust SigNoz-issued tokens.
  • Users can be silently redirected to phishing or malware sites during the SSO login process.
  • Attackers may steal access and refresh tokens for arbitrary SigNoz users.
  • Compromised tokens could allow full account takeover for impacted users.
  • Attackers can access or modify observability data and dashboards associated with victim accounts.
  • Malicious actors might pivot into other integrated systems that trust SigNoz-issued tokens.
  • Users can be silently redirected to phishing or malware sites during the SSO login process.

General mitigation guidance

Effective mitigation focuses on constraining and validating redirect behavior in authentication flows:

  • Apply a strict allowlist for post-authentication redirect URIs and paths, rejecting or normalizing any value that does not match known internal origins.
  • Treat all client-supplied redirect parameters as untrusted and avoid embedding them directly into token-bearing URLs.
  • Decouple token issuance from user-controlled redirects, using server-side session establishment and opaque state values rather than passing tokens via query parameters to arbitrary destinations.
  • Enforce defense-in-depth by limiting token scope, lifetime, and audience so that compromised tokens have reduced impact.
  • Prefer secure-by-default SSO integrations where redirect targets are configured centrally and not dynamically influenced by runtime user input.

Effective mitigation focuses on constraining and validating redirect behavior in authentication flows:

  • Apply a strict allowlist for post-authentication redirect URIs and paths, rejecting or normalizing any value that does not match known internal origins.
  • Treat all client-supplied redirect parameters as untrusted and avoid embedding them directly into token-bearing URLs.
  • Decouple token issuance from user-controlled redirects, using server-side session establishment and opaque state values rather than passing tokens via query parameters to arbitrary destinations.
  • Enforce defense-in-depth by limiting token scope, lifetime, and audience so that compromised tokens have reduced impact.
  • Prefer secure-by-default SSO integrations where redirect targets are configured centrally and not dynamically influenced by runtime user input.

How teams detect issues like this

Organizations can detect similar issues by:

  • Using static analysis or code scanning rules to flag uses of request parameters in redirect APIs, especially within authentication or SSO modules.
  • Reviewing SSO and OAuth/OIDC integration code for any unvalidated redirect_uri, ref, or next parameters that control navigation or token delivery.
  • Implementing security review checklists that mandate whitelist-based redirect handling for login flows and forbid token exposure in client-controlled redirects.
  • Analyzing web server logs for redirects to external domains initiated from authentication endpoints, looking for patterns involving shared parameters or abnormal destinations.
  • Enforcing policies in CI/CD that require manual review of changes to authentication and SSO routing logic, optionally supported by automated web security testing tools capable of identifying open redirects.

Organizations can detect similar issues by:

  • Using static analysis or code scanning rules to flag uses of request parameters in redirect APIs, especially within authentication or SSO modules.
  • Reviewing SSO and OAuth/OIDC integration code for any unvalidated redirect_uri, ref, or next parameters that control navigation or token delivery.
  • Implementing security review checklists that mandate whitelist-based redirect handling for login flows and forbid token exposure in client-controlled redirects.
  • Analyzing web server logs for redirects to external domains initiated from authentication endpoints, looking for patterns involving shared parameters or abnormal destinations.
  • Enforcing policies in CI/CD that require manual review of changes to authentication and SSO routing logic, optionally supported by automated web security testing tools capable of identifying open redirects.

Sources

SourceLink
github.comgithub.com/SigNoz/signoz/commit/253ca7dd7eb4f7a32a694c249eb0d5d0804d5619
github.comgithub.com/SigNoz/signoz/issues/11746
github.comgithub.com/SigNoz/signoz/pull/12172
github.comgithub.com/SigNoz/signoz/releases/tag/v0.134.0
vulncheck.comvulncheck.com/advisories/signoz-sso-oauth-state-manipulation-session-token-theft

Frequently asked

questions

It is broadly exploitable because attackers do not need prior access or authentication; they simply need to craft a malicious SSO login link and convince a user of a vulnerable SigNoz instance to use it.

No authenticated access is required. The attacker only needs network access to the SigNoz web interface and the ability to deliver crafted URLs to potential victims (for example, via email or chat).

The safest pattern is to enforce strict, server-side validation of all redirect destinations in authentication flows using an allowlist of trusted origins and paths, and to avoid sending tokens or sensitive session data to any location influenced by user-controlled parameters.

Treat all redirect parameters as untrusted input, apply allowlist-based validation, keep token issuance and navigation logic tightly controlled on the server, and include explicit checks for open redirects and token exposure in security design reviews for any new authentication, SSO, or OAuth/OIDC integrations.

It is broadly exploitable because attackers do not need prior access or authentication; they simply need to craft a malicious SSO login link and convince a user of a vulnerable SigNoz instance to use it.

No authenticated access is required. The attacker only needs network access to the SigNoz web interface and the ability to deliver crafted URLs to potential victims (for example, via email or chat).

The safest pattern is to enforce strict, server-side validation of all redirect destinations in authentication flows using an allowlist of trusted origins and paths, and to avoid sending tokens or sensitive session data to any location influenced by user-controlled parameters.

Treat all redirect parameters as untrusted input, apply allowlist-based validation, keep token issuance and navigation logic tightly controlled on the server, and include explicit checks for open redirects and token exposure in security design reviews for any new authentication, SSO, or OAuth/OIDC integrations.

Find out what's already
exploitable in your codebase.