docker-build

Warn

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides examples in SKILL.md and references/01-build-command-reference.md of using subshell expansion to pass build arguments from a file: docker build --build-arg $(cat .env.build | xargs). This pattern can be exploited for arbitrary command execution on the host system if the .env.build file content is untrusted.\n- [EXTERNAL_DOWNLOADS]: The skill includes instructions for building images directly from remote Git repositories (e.g., https://github.com/user/repo.git#main). This involves fetching and potentially executing code from external sources during the build process.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via processed Dockerfiles and build contexts.\n
  • Ingestion points: Local files, stdin, and remote Git repositories (found in SKILL.md).\n
  • Boundary markers: None identified. No delimiters or instructions are used to isolate untrusted content.\n
  • Capability inventory: docker build (local command execution), docker push (network access), and image management commands.\n
  • Sanitization: No input validation or instruction filtering for external content is present.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 22, 2026, 10:25 AM
Security Audit — agent-trust-hub — docker-build