自动化接入 · API / MCP EN

技术说明 · T1195.001

Compromise Software Dependencies and Development Tools

Adversaries may manipulate software dependencies and development tools prior to receipt by a final consumer for the purpose of data or system compromise. Applications often depend on external software to function properly. Popular open source projects that are used as dependencies in many applications, such as pip and NPM packages, may be targeted as a means to add malicious code to users of the dependency. This may also include abandoned packages, which in some cases could be re-registered by threat actors after being removed by the original owner. Adversaries may also employ "typosquatting" or name-confusion by choosing names similar to existing popular libraries or packages in order to deceive a user. Additionally, CI/CD pipeline components, such as GitHub Actions, may be targeted in order to gain access to the building, testing, and deployment cycles of an application. By adding malicious code into a GitHub action, a threat actor may be able to collect runtime credentials (e.g., via Proc Filesystem) or insert further malicious components into the build pipelines for a second-order supply chain compromise. As GitHub Actions are often dependent on other GitHub Actions, threat actors may be able to infect a large number of repositories via the compromise of a single Action. Adversaries may also target the configuration and instruction files consumed by integrated development environments and AI coding agents. Project files such as tasks.json may execute commands when a developer opens or builds a project, while agent skill files (SKILL.md) supply an agent with instructions and resources that it loads on demand and runs with the permissions of the agent process. To encourage downloads, adversaries may publish these files under benign-appearing or typo-squatted names, upload unmodified copies of the genuine skills first so the listings accumulate installs before the content behind them is weaponized, or may place the malicious instructions in a secondary document that the reviewed entry point references. Because these artifacts are natural language or configuration rather than executable code, they may pass registry screening and signature-based scanning. Targeting may be specific to a desired victim set or may be distributed to a broad set of consumers but only move on to additional tactics on specific victims.

ATT&CK T1195.001 platforms — 状态 有效 版本 20.1

Adversaries may manipulate software dependencies and development tools prior to receipt by a final consumer for the purpose of data or system compromise. Applications often depend on external software to function properly. Popular open source projects that are used as dependencies in many applications, such as pip and NPM packages, may be targeted as a means to add malicious code to users of the dependency.123 This may also include abandoned packages, which in some cases could be re-registered by threat actors after being removed by the original owner.4 Adversaries may also employ "typosquatting" or name-confusion by choosing names similar to existing popular libraries or packages in order to deceive a user.567 Additionally, CI/CD pipeline components, such as GitHub Actions, may be targeted in order to gain access to the building, testing, and deployment cycles of an application.8 By adding malicious code into a GitHub action, a threat actor may be able to collect runtime credentials (e.g., via Proc Filesystem) or insert further malicious components into the build pipelines for a second-order supply chain compromise.9 As GitHub Actions are often dependent on other GitHub Actions, threat actors may be able to infect a large number of repositories via the compromise of a single Action.10 Adversaries may also target the configuration and instruction files consumed by integrated development environments and AI coding agents. Project files such as tasks.json may execute commands when a developer opens or builds a project, while agent skill files (SKILL.md) supply an agent with instructions and resources that it loads on demand and runs with the permissions of the agent process.11121314 To encourage downloads, adversaries may publish these files under benign-appearing or typo-squatted names, upload unmodified copies of the genuine skills first so the listings accumulate installs before the content behind them is weaponized, or may place the malicious instructions in a secondary document that the reviewed entry point references.1516 Because these artifacts are natural language or configuration rather than executable code, they may pass registry screening and signature-based scanning.1718 Targeting may be specific to a desired victim set or may be distributed to a broad set of consumers but only move on to additional tactics on specific victims.

  1. Trendmicro NPM Compromise
  2. Bitdefender NPM Repositories Compromised 2021
  3. MANDVI Malicious npm and PyPI Packages Disguised
  4. The Hacker News PyPi Revival Hijack 2024
  5. Ahmed Backdoors in Python and NPM Packages
  6. Meyer PyPI Supply Chain Attack Uncovered
  7. Checkmarx-oss-seo
  8. Unit 42 Palo Alto GitHub Actions Supply Chain Attack 2025
  9. OWASP CICD-SEC-4
  10. Palo Alto Networks GitHub Actions Worm 2023
  11. Microsoft - Agent Skills Docs
  12. Google Agent Files May 2026
  13. CrowdStrike Threat Hunting Report 2026
  14. Microsoft Contagious Interview March 2026
  15. Zenity Skill Supply Chain August 2026
  16. ESET Agent Skills March 2026
  17. ESET Agent Skills March 2026
  18. TrendMicro - OpenClaw Skills Distribute ATMOS

使用该技术的组织

组织名称国家
grp_g1056TeamPCP—
grp_g1057ShinyHunters—
grp_g0082APT38DPRK

断言

尚无指向该技术的公开断言。

记录自动入库·未人工复核；置信度与有效期按引用读取。