TeamPCP Ransomware Shift Raises Threat Despite Slower Attacks

▼ Summary
– TeamPCP’s recent string of supply chain attacks has paused, with no new open-source project compromises reported in the three days since their last PyPI upload.
– The group has partnered with a new ransomware-as-a-service operation called Vect, shifting focus from new attacks to monetizing previously stolen credentials.
– TeamPCP rapidly evolved from targeting misconfigured servers in 2024 to executing sophisticated, automated supply chain attacks by 2026.
– Their attacks have a wide ripple effect, with hundreds of repositories and packages automatically pulling malicious code through dependencies.
– Security experts warn against automatically updating to the latest package versions and recommend pinning dependencies to specific cryptographic hashes.
A recent lull in TeamPCP’s aggressive supply chain campaign does not signal a retreat. Instead, security analysts warn the group has pivoted to a more dangerous phase, partnering with a new ransomware-as-a-service (RaaS) operation to monetize the vast trove of credentials stolen during its recent spree. After a relentless series of attacks through March, the group has not released a new poisoned open-source package in three days, a strategic shift experts link directly to its new alliance.
This operational pause coincides with an announcement from a fledgling RaaS outfit called Vect on the cybercriminal forum BreachForum. Vect declared a partnership with TeamPCP and a plan to enlist all forum members as affiliates. “Together, we are ready to deploy ransomware across all affected companies that got hit by these attacks,” the group boasted, promising even larger future operations. According to SANS instructor Kenneth Hartman, this threat is already materializing, with a confirmed Vect ransomware deployment using credentials sourced by TeamPCP.
The group’s evolution has been alarmingly rapid. Emerging in 2024 with attacks on misconfigured cloud services, TeamPCP built automated supply chain attack capabilities by 2025. It later deployed the novel CanisterWorm, using decentralized infrastructure for command and control. The March 2026 campaign represented a peak, cascading through five vendor ecosystems from a single stolen credential. Throughout these attacks, the group demonstrated remarkable adaptability, quickly pivoting across target software and evolving its delivery techniques from Base64 encoding to sophisticated steganography.
Hartman emphasizes that the current hiatus in new package compromises should not breed complacency. TeamPCP has stated its intention to persist long-term, and its estimated 300 GB cache of stolen credentials could fuel future attacks at any moment. The pause may also reflect increased defensive vigilance, such as PyPI’s recent quarantine of two TeamPCP campaigns, potentially raising the group’s operational costs on that platform.
The downstream impact of these attacks has been significant. Analysis by GitGuardian found that 474 public repositories executed malicious code from a compromised CI/CD component, while 1,750 Python packages were configured to automatically pull poisoned versions of a compromised library. These figures are considered conservative, as they exclude private repositories and deeper dependency chains. Researchers stress that pinning dependencies to cryptographic hashes, rather than automatically updating to the latest release, is a critical defensive measure.
This incident underscores a dangerous practice in modern development. The reflexive push to use the newest software version the moment it drops creates a massive vulnerability. If a CI/CD pipeline automatically pulls the latest release without a quarantine period, it effectively automates its own compromise. The security community advises letting other infrastructure test new releases for supply chain threats first. For organizations investigating potential exposure, sharing the SHA256 digests of known malicious packages remains the only definitive method to check for compromise.
(Source: Help Net Security)


