Namastex.ai npm Packages Hit with TeamPCP-Style CanisterWorm Malware

Key Takeaways

• Malicious npm packages linked to Namastex.ai (@automagik/genie, pgserve) were compromised with CanisterWorm-style malware.
• The malware executes at install-time via postinstall hooks to harvest sensitive developer secrets, cloud credentials, and crypto-wallet data.
• Exfiltration occurs via a conventional webhook and an Internet Computer Protocol (ICP) canister endpoint.
• The payload includes self-propagation logic to steal npm tokens and PyPI credentials to infect and republish other packages.

Affected Systems

• Developer workstations
• CI/CD pipelines
• npm ecosystem
• PyPI ecosystem
• Linux/macOS environments (implied by targeted file paths)

Attack Chain

The attack begins when a developer installs a compromised npm package, triggering malicious code execution via a postinstall hook. The malware harvests sensitive environment variables, cloud credentials, SSH keys, and crypto-wallet data from the local system. Stolen data is encrypted using a bundled RSA public key and exfiltrated to a webhook and an ICP canister endpoint. Finally, the malware attempts to self-propagate by extracting npm and PyPI tokens from the victim machine to inject and republish malicious packages.

Detection Availability

• YARA Rules: No
• Sigma Rules: No
• Snort/Suricata Rules: No
• KQL Queries: No
• Splunk SPL Queries: No
• EQL Queries: No
• Other Detection Logic: No

The article does not provide specific detection rules, but lists file hashes, domains, and distinctive strings for threat hunting.

Detection Engineering Assessment

EDR Visibility: Medium — EDR can monitor process executions (like node.js reading sensitive files such as ~/.aws/credentials), but might miss the initial npm install context without specific supply chain monitoring. Network Visibility: Medium — Network monitoring can detect connections to the known malicious webhook and ICP canister domains, though the traffic is likely HTTPS encrypted. Detection Difficulty: Moderate — Detecting the specific exfiltration endpoints is straightforward, but distinguishing malicious npm postinstall activity from legitimate build scripts requires behavioral baselining.

Required Log Sources

• Process Creation (Event ID 4688 / Sysmon 1)
• File Access (Sysmon 11)
• DNS Query (Sysmon 22)
• Network Connection (Sysmon 3)

Hunting Hypotheses

Control Gaps

• Lack of egress filtering for developer workstations
• Implicit trust in npm package postinstall scripts

Key Behavioral Indicators

• Node.js executing dist/env-compat.cjs
• Presence of X-Session-ID and X-Request-Signature headers in unexpected outbound traffic from dev tools
• sqlite3 accessing Chrome Login Data from a Node context

False Positive Assessment

• Low

Recommendations

Immediate Mitigation

• Block and remove the identified malicious package versions (@automagik/genie, pgserve, etc.) from all environments.
• Rotate all npm tokens, GitHub tokens, cloud credentials, and SSH keys present on affected systems.

Infrastructure Hardening

• Implement strict egress filtering on CI/CD pipelines and developer workstations to block unknown domains.
• Disable npm postinstall scripts globally (npm config set ignore-scripts true) where not strictly required.

User Protection

• Deploy endpoint monitoring to alert on unauthorized access to sensitive files (e.g., ~/.aws, ~/.ssh).
• Use hardware security keys (FIDO2) for critical infrastructure access to mitigate credential theft.

Security Awareness

• Educate developers on the risks of typosquatting and supply chain attacks.
• Establish a policy for reviewing new dependencies and their updates before integration.

MITRE ATT&CK Mapping

• T1195.001 - Compromise Software Dependencies and Development Tools
• T1059.007 - Command and Scripting Interpreter: JavaScript
• T1552.001 - Credentials In Files
• T1552.004 - Private Keys
• T1555.003 - Credentials from Web Browsers
• T1048 - Exfiltration Over Alternative Protocol

Additional IOCs

• Domains:
  • cjn37-uyaaa-aaaac-qgnva-cai[.]raw[.]icp0[.]io - ICP canister domain for exfiltration
• Urls:
  • hxxps://telemetry[.]api-monitor[.]com/v1/drop - Secondary webhook exfiltration URL
• File Paths:
  • ~/.npmrc - Targeted for credential theft
  • ~/.ssh/ - Targeted for SSH key theft
  • ~/.git-credentials - Targeted for credential theft
  • ~/.netrc - Targeted for credential theft
  • ~/.aws/ - Targeted for cloud credential theft
  • ~/.config/gh/hosts.yml - Targeted for GitHub credential theft
  • ~/.kube/config - Targeted for Kubernetes credential theft
  • ~/.docker/config.json - Targeted for Docker credential theft
  • /proc/*/environ - Targeted for environment variable theft on Linux
  • dist/env-compat.cjs - Malware payload file
  • dist/public.pem - Bundled RSA public key
• Command Lines:
  • Purpose: Execute malicious payload via postinstall hook | Tools: node | Stage: Execution | node dist/env-compat.cjs
• Other:
  • 87259b0d1d017ad8b8daa7c177c2d9f0940e457f8dd1ab3abab3681e433ca88e - DER SHA-256 fingerprint of RSA public key
  • cjn37-uyaaa-aaaac-qgnva-cai - Hardcoded ICP canister ID
  • pkg-telemetry - Distinctive string in malware
  • dist-propagation-report - Distinctive string in malware
  • pypi-pth-exfil - Distinctive string in malware
  • X-Session-ID - Distinctive HTTP header string used by malware
  • X-Request-Signature - Distinctive HTTP header string used by malware
  • @automagik/genie - Compromised npm package
  • pgserve - Compromised npm package
  • @fairwords/websocket - Compromised npm package
  • @fairwords/loopback-connector-es - Compromised npm package
  • @openwebconcept/design-tokens - Compromised npm package
  • @openwebconcept/theme-owc - Compromised npm package