New PhantomRaven NPM Attack Wave Steals Dev Data via 88 Packages

Why It Matters

The attack compromises developer credentials and CI/CD pipelines, enabling attackers to infiltrate production environments and steal sensitive code. Its reliance on AI‑generated names and RDD evades traditional package scans, raising supply‑chain risk for the broader software ecosystem.

Key Takeaways

• 88 malicious packages released via 50 disposable npm accounts
• Attack uses slopsquatting to mimic popular libraries
• Remote Dynamic Dependencies fetch code from external URLs
• Malware exfiltrates .gitconfig, .npmrc, CI/CD tokens
• 81 of 88 packages remain publicly available on npm

Pulse Analysis

The npm ecosystem has become a prime target for supply‑chain attacks, and the latest PhantomRaven wave underscores how attackers adapt to developer workflows. By leveraging "slopsquatting"—a technique that mimics well‑known libraries such as Babel and GraphQL Codegen—threat actors increase the likelihood of accidental installs. Coupled with Remote Dynamic Dependencies, the malicious code resides on attacker‑controlled servers, bypassing static analysis that only inspects package contents. This approach reflects a broader trend where adversaries exploit the trust developers place in package managers and AI‑generated suggestions.

Technically, the PhantomRaven payload is streamlined yet potent. Once a compromised package is installed, the RDD mechanism pulls a remote script that harvests sensitive files like .gitconfig and .npmrc, as well as CI/CD tokens from GitHub, GitLab, Jenkins, and CircleCI. System fingerprints—including IP address, hostname, OS, and Node version—are collected to tailor exfiltration. Data is transmitted via HTTP GET, POST, or WebSocket to domains featuring the word "artifact" on unsecured Amazon EC2 instances. The consistency across four observed waves, with 257 of 259 code lines unchanged, demonstrates a low‑maintenance, high‑return operation.

Mitigation requires a blend of policy and tooling. Organizations should enforce strict provenance checks, favoring packages from verified publishers and scrutinizing any dependency that references external URLs. Automated scanning tools need to incorporate RDD detection, while developers must resist copying AI‑suggested package names without validation. Regular credential rotation and token scoping further reduce the impact of potential breaches. As supply‑chain threats evolve, a proactive stance—combining education, robust tooling, and continuous monitoring—will be essential to safeguard the software development lifecycle.