docker-compose-stacks
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided service and stack names which are then used as arguments for Docker management tools.
- Ingestion points: Service names (e.g., 'Grafana', 'AdGuard') and stack file identifiers (e.g., 'compose.monitoring.yml') are provided by the user via chat input.
- Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are present to isolate the interpolated user data.
- Capability inventory: The skill uses tools capable of making significant system changes, including
homelab_composeUp,homelab_composeDown, andhomelab_serviceRestart(SKILL.md). - Sanitization: No sanitization, validation, or escaping of user-supplied strings is documented in the skill's instructions before they are passed to the tools.
Audit Metadata