Loading malware…
DUSTMAKER is a cross-platform JavaScript credential stealer used by the financially motivated threat actor TeamPCP, also tracked as UNC6780 and Altered Spider. Active since April 2026, it succeeded the group’s SANDCLOCK credential stealer and is optimized for CI/CD pipelines, AI development tooling, and developer environments. DUSTMAKER steals credentials to support TeamPCP extortion operations and has been associated with software supply-chain compromises affecting open-source development ecosystems. Unlike SANDCLOCK, DUSTMAKER is not reported to include container-escape functionality. Variants hide malicious content within AI coding-assistant and IDE workspaces and can abuse workspace configuration to induce AI assistants to execute scripts during normal development activity. It also employs prompt-injection content intended to hinder LLM-assisted security analysis.
2 distinct threat actors Mallory has associated with this family across public reporting. Open in Mallory to see the full evidence chain and overlapping campaigns.
The report also describes DUSTMAKER, tied to UNC6780 or TeamPCP, which targeted AI development tools and CI/CD systems.
DUSTMAKER is a successor to the SANDCLOCK credential stealer in TeamPCP operations and is a cross-platform JavaScript payload optimized for CI/CD pipelines.
5 distinct techniques documented for this family, organized by ATT&CK tactic.
DUSTMAKER, tied to UNC6780 or TeamPCP, targeted AI development tools and CI/CD systems... a GitHub Actions workflow weakness similarly underlines how compromised automation can expose OIDC tokens and push malicious changes downstream.
DUSTMAKER... could use malicious configuration files to influence an assistant into running scripts during routine development work... File name setup.mjs [was] capable of being executed through malicious workspace configuration.
Hidden AI coding assistant workspace directory used by DUSTMAKER... Hidden IDE workspace directory used to blend malicious files into projects.
“The threat actor leveraged an AI coding chatbot, a prompt, and a set of agent instructions to plan, build, and execute a mass credential harvesting campaign in less than six hours.”
The agents actively searched server-side systems, identified weaknesses, and carried out targeted actions against online infrastructure... [The Recon] dashboard was designed to organize, validate, and manage more than 23,800 stolen secrets in real time, including API keys connected to cloud and AI services.
“The threat actor leveraged an AI coding chatbot, a prompt, and a set of agent instructions to plan, build, and execute a mass credential harvesting campaign in less than six hours.”
5 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.