Skip to main content
Vulnerability Database/CVE-2026-97877

CVE-2026-97877: StarTraining JWT Auth Bypass Vulnerability

CVE-2026-97877 is an authentication bypass flaw in zhistaredu StarTraining up to version 3.8.1 affecting JWT token handling. Attackers can exploit hard-coded credentials remotely. This article covers technical details, affected versions, impact assessment, and mitigation strategies.

Published:

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

CVE-2026-97877 Overview

CVE-2026-97877 affects zhistaredu StarTraining versions up to 3.8.1. The vulnerability resides in the UserLoginService.createToken function, which handles JSON Web Token (JWT) generation. A hard-coded secret stored in application.yml allows attackers to forge valid JWT tokens by manipulating the user_id and company_id arguments [CWE-255]. The flaw is exploitable remotely without authentication or user interaction. The exploit has been publicly disclosed, and the vendor did not respond to disclosure attempts.

Critical Impact

Remote attackers can forge arbitrary JWT tokens to impersonate any user or company tenant within StarTraining deployments.

Affected Products

  • zhistaredu StarTraining versions through 3.8.1
  • Component: JWT Token Handler (UserLoginService.createToken)
  • Configuration file: application.yml

Discovery Timeline

  • 2026-09-25 - CVE-2026-97877 published to the National Vulnerability Database (NVD)
  • 2026-09-28 - Last updated in NVD database

Technical Details for CVE-2026-97877

Vulnerability Analysis

StarTraining uses JWTs for session management. The signing secret used by the UserLoginService.createToken function is embedded as a hard-coded value within the application.yml configuration file shipped with the application. Because the secret ships with every installation, any attacker who inspects the public source or a deployed binary can recover it. With the secret in hand, an attacker can craft signed tokens that the server accepts as legitimate.

The forged tokens can specify arbitrary user_id and company_id claim values. The backend trusts these claims because signature verification succeeds against the shared secret. This provides a direct path to account impersonation and cross-tenant access without ever interacting with the login flow.

Root Cause

The root cause is use of a hard-coded cryptographic secret [CWE-255]. The JWT signing key is not generated per deployment and is not rotated. All instances of StarTraining up to version 3.8.1 share the same signing material, which breaks the trust boundary that JWT signatures are meant to enforce.

Attack Vector

An unauthenticated remote attacker retrieves the hard-coded secret from the published source or a reachable installation. The attacker then generates a JWT with chosen user_id and company_id claims, signed with the known secret. The attacker submits the forged token to any protected API endpoint and is granted the privileges associated with the claimed identity. No prior credentials, phishing, or user interaction are required.

See the GitHub Security Advisory ST-VULN-001 for proof-of-concept details.

Detect Identity Attacks the Moment Authentication Fails

Singularity™ Identity surfaces credential misuse, lateral movement, and privilege escalation in real time, so you can contain the breach before it spreads.

Stop Identity Attacks

Detection Methods for CVE-2026-97877

Indicators of Compromise

  • JWT access tokens presented to StarTraining APIs that were not issued by the login flow in authentication logs.
  • API requests containing user_id or company_id claims that do not correspond to recent login events.
  • Session activity for user or tenant identifiers that never completed the UserLoginService.createToken code path.

Detection Strategies

  • Correlate JWT iat and jti claims from inbound requests against server-side issuance records to flag tokens with no matching issuance event.
  • Alert on authenticated API calls originating from source IP addresses that never performed a successful login handshake.
  • Monitor for abrupt horizontal access patterns where a single session touches multiple company_id tenants.

Monitoring Recommendations

  • Enable verbose logging on the UserLoginService component and ingest it into a centralized analytics platform for correlation.
  • Baseline normal JWT issuance volume and alert on anomalies in token acceptance versus token issuance ratios.
  • Review web application firewall (WAF) and reverse proxy logs for repeated requests bearing structurally valid but previously unseen JWTs.

How to Mitigate CVE-2026-97877

Immediate Actions Required

  • Rotate the JWT signing secret in application.yml to a high-entropy value unique to each deployment, and invalidate all currently issued tokens.
  • Restrict network exposure of StarTraining APIs to trusted networks or require an authenticated reverse proxy in front of the application.
  • Audit recent authentication and API access logs for signs of forged-token activity across all tenants.

Patch Information

No vendor patch is available. The vendor was contacted but did not respond to the disclosure. Operators should treat version 3.8.1 and earlier as unpatched and apply the workarounds below until an upstream fix is published. Refer to VulDB CVE-2026-97877 for ongoing tracking.

Workarounds

  • Replace the hard-coded secret in application.yml with an externally injected value sourced from an environment variable or secrets manager.
  • Enforce short JWT expiration windows and implement server-side token revocation to limit the useful lifetime of any forged token.
  • Add server-side validation that cross-checks the user_id and company_id claims against an authoritative session store before granting access.
bash
# Configuration example: externalize the JWT secret
# In application.yml, replace the hard-coded value with an environment reference
# jwt:
#   secret: ${JWT_SIGNING_SECRET}

# Generate a strong per-deployment secret and export it before starting the service
export JWT_SIGNING_SECRET="$(openssl rand -base64 48)"

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

Experience the Most Advanced Cybersecurity Platform

See how the world’s most intelligent, autonomous cybersecurity platform can protect your organization today and into the future.