150,000+
Cloud security firm Wiz has identified a critical supply chain risk within the Visual Studio Code (VSCode) and OpenVSX extension marketplaces, which are used by millions of software developers worldwide. The research team discovered that publishers of more than 100 extensions had accidentally leaked their access tokens. This exposure created a scenario where a threat actor could have seized control of these extensions, pushed malicious updates, and potentially distributed malware to over 150,000 users. The investigation also unearthed over 550 other exposed secrets (e.g., API keys, credentials) within more than 500 extensions, revealing a widespread and systemic issue of poor secrets management in the software development lifecycle.
The core of the issue lies in the leakage of publisher access tokens. These tokens are essentially the keys to the kingdom for an extension; whoever possesses one can publish new versions, modify the extension's code, and change its description. Wiz researchers found over 100 such tokens that were publicly exposed, often because they were accidentally committed to public GitHub repositories.
A threat actor with one of these tokens could have performed a devastating supply chain attack:
Beyond the publisher tokens, the researchers found an additional 550+ exposed secrets within the code of over 500 different extensions. These included API keys and credentials for services like OpenAI, Anthropic, Amazon Web Services (AWS), and MongoDB. This secondary finding highlights a broader culture of insecure coding practices and presents an additional attack surface, allowing attackers to compromise developers' cloud infrastructure and services.
code.exe) making unexpected network connections or accessing sensitive files.Treating third-party extensions as part of the software supply chain and vetting them for security is a critical mitigation.
Implementing secure coding practices, such as not hardcoding secrets, is a fundamental configuration control for developers.
Mapped D3FEND Techniques:
Providing developers with clear guidance and tools for secrets management and secure coding.
To prevent the inadvertent leakage of secrets as seen in the VSCode marketplace, organizations must integrate automated static analysis security testing (SAST) and secrets scanning directly into their CI/CD pipelines. Before any code is merged into a main branch or deployed, these tools should automatically scan the entire codebase for hardcoded credentials, API keys, and access tokens. If a secret is found, the build should fail automatically, preventing the secret from ever being committed to a shared repository like GitHub. This 'shift-left' approach moves security into the earliest stages of development and is the most effective way to prevent this specific type of supply chain risk at its source.
Developers must adopt secure coding practices that eliminate hardcoded secrets entirely. Instead of placing tokens in code or configuration files, use a dedicated secrets management service (e.g., HashiCorp Vault, AWS Secrets Manager, Azure Key Vault). The application should be configured to dynamically fetch secrets from this vault at runtime using a secure authentication method (like IAM roles for cloud workloads). This ensures that secrets are never present in the source code, and therefore cannot be leaked through an accidental commit to a public repository. This is a fundamental change in application architecture that is essential for modern, secure software development.

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.
Help others stay informed about cybersecurity threats