docker

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various administrative commands using the docker and docker compose binaries to manage the host's container environment, including system-level pruning and network configuration.
  • [REMOTE_CODE_EXECUTION]: The skill uses docker run, docker exec, and docker build which allow the agent to execute arbitrary code or processes within a containerized environment. This includes pulling and running external images from Docker Hub.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external container outputs.
  • Ingestion points: Reads data through docker logs, docker ps, and docker inspect (SKILL.md).
  • Boundary markers: No boundary markers or specific instructions are provided to ignore malicious content within the logs.
  • Capability inventory: The skill possesses significant capabilities including container execution (docker exec), container creation (docker run), and image building (docker build) (SKILL.md).
  • Sanitization: No sanitization or filtering of container output or log data is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 05:13 AM