Artifact Poisoning

Artifact poisoning is a type of software supply chain attack where a malicious actor modifies or injects malicious code into build outputs, such as binaries, containers, libraries, or software packages, to compromise downstream users or environments.

Software artifacts are often trusted blindly once built and published. If poisoned, they can silently compromise users, introduce backdoors, or act as a vector for malware. This makes artifact poisoning a high-impact, low-visibility threat, particularly in environments relying on automated CI/CD workflows and artifact reuse.

Common techniques include:

• Compromising the build pipeline to inject malicious code
• Altering package contents before publishing to artifact repositories
• Hijacking third-party libraries with poisoned versions
• Replacing legitimate artifacts with trojanized copies on the wire or in storage

Attackers often exploit weak controls over build agents, repositories, or access tokens, using these to modify artifacts before they're delivered or consumed.

• Preserves Product Integrity: Ensures shipped software matches what was intended and reviewed.
• Reduces Legal and Regulatory Risk: Meets growing expectations under SBOM, EO 14028, and NIST SSDF guidelines.
• Prevents Customer Impact: Stops the propagation of malware or exploitable code to downstream environments, thereby minimizing the potential for customer impact.
• Strengthens Supply Chain Trust:Supports tamper-proof software delivery practices for both internal and external stakeholders, ensuring transparency and integrity.