docker-expert
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes Dockerfiles and Compose files in the local environment to provide recommendations and perform builds, creating a potential surface for indirect prompt injection.
- Ingestion points: Reads local files identified by find commands (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Performs docker build, docker run, and docker exec (SKILL.md).
- Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill utilizes shell commands for environment discovery and container lifecycle management.
- Evidence: Executes docker version, docker info, docker build, and docker ps to assess and validate the Docker environment (SKILL.md).
Audit Metadata