Checklist Outlines Defenses Against Software Supply-Chain Attacks

Summary
A practical checklist recommends defenses against software supply-chain attacks, including package-age policies, pinned CI dependencies, scoped agent permissions, prompt-injection safeguards, and audit logging.
Key points
- The checklist covers dependency management, access controls, CI/CD, containers, developer environments, agentic tools, and monitoring and response.
- Use package-age policies to delay installation of newly published malicious releases.
- Pin third-party pipeline actions to commit SHAs rather than movable tags.
- Treat MCP servers as third-party dependencies and limit their permissions.
- Protect AI-powered pipelines from prompt injection through inputs such as issue titles and pull-request descriptions.
- Capture audit logs for registry pushes, CI runs, and token use; valid live sessions may not generate useful signals in standard authentication logs.
- The full checklist also includes lockfile enforcement, scoped tokens, phishing-resistant MFA, container hardening, and incident-response playbooks.
Article Details
- Defense Focus
- Prevent and detect software supply chain compromises across dependency installation, CI/CD pipelines, developer environments, and agentic tooling.
- Detection Methods
- Review audit logs for package registry pushes, CI runs, and token usage, including activity performed through valid sessions.
- Compare audit-log activity with standard authentication logs, which the article says did not reveal several compromises.
- Data Sources
- Package registry push audit logs
- CI run audit logs
- Token-usage audit logs
- Authentication logs
- Platforms
- Package registries
- CI/CD pipelines
- Developer machines
- IDE extensions
- Browser plugins
- AI-powered pipelines
- Defensive Actions
- Enforce a package-age policy before dependencies can install.
- Pin pipeline dependencies, including third-party actions, to commit SHAs and enforce lockfiles.
- Use scoped tokens and phishing-resistant MFA.
- Treat MCP servers as third-party dependencies and scope agent permissions.
- Harden container images and control IDE extensions and browser plugins.
- Guard against prompt injection in AI-powered pipelines.
- Capture audit logs for registry pushes, CI runs, and token usage.
- Maintain incident-response playbooks for supply chain compromises.
MITRE ATT&CK
Products
Axiostj-actions GitHub Action was compromised and pulled malicious code into thousands of pipelines, and the axios maintainer account was hijacked and used to distribute a RAT. Malicious releases also hit Zapier, ENSBitwarden CLIhijacked and used to distribute a RAT. Malicious releases also hit Zapier, ENS Domains, PostHog, and Bitwarden CLI.Chalkare having a moment. Open-source malware detections jumped 73% in 2025. In the past year, the debug and chalk packages were backdoored, the tj-actions GitHub Action was compromised and pulled malicious code intoDebugattacks are having a moment. Open-source malware detections jumped 73% in 2025. In the past year, the debug and chalk packages were backdoored, the tj-actions GitHub Action was compromised and pulled malicious codeENS Domainsmaintainer account was hijacked and used to distribute a RAT. Malicious releases also hit Zapier, ENS Domains, PostHog, and Bitwarden CLI.MastraThe pattern across many of the recent big attacks like axios, chalk and debug, and Mastra is that malicious releases of trusted packages started executing within hours of publication. Install-time controls are what buyPostHogaccount was hijacked and used to distribute a RAT. Malicious releases also hit Zapier, ENS Domains, PostHog, and Bitwarden CLI.tj-actionsdetections jumped 73% in 2025. In the past year, the debug and chalk packages were backdoored, the tj-actions GitHub Action was compromised and pulled malicious code into thousands of pipelines, and the axiosZapierand the axios maintainer account was hijacked and used to distribute a RAT. Malicious releases also hit Zapier, ENS Domains, PostHog, and Bitwarden CLI.
