docker-swarm

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and scripts for executing administrative Docker commands, such as docker swarm init, docker node promote, and docker stack deploy. These operations typically require root or docker-group privileges to manage system-level resources and container orchestration.
  • [INDIRECT_PROMPT_INJECTION]: The skill handles external configuration data and templates which could serve as a vector for indirect injection if provided by an untrusted source.
  • Ingestion points: Processes stack.yaml and nginx.conf files during deployment tasks (e.g., docker stack deploy -c stack.yaml).
  • Boundary markers: The instructions do not define delimiters or warnings to prevent the agent from interpreting instructions embedded within these configuration files.
  • Capability inventory: The skill has the capability to execute shell commands and manage cluster state via the docker CLI and the included scripts/swarm-init.sh script.
  • Sanitization: There is no evidence of validation, escaping, or filtering of the content within the configuration files or environment variables before they are used in deployment commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:51 PM