skills/bmad-code-org/bmad-module-creative-intelligence-suite/bmad-cis-agent-presentation-master/Gen Agent Trust Hub
bmad-cis-agent-presentation-master
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes a local script 'resolve_customization.py' via 'uv run' during activation to process configuration layers. This script is located within the project's '_bmad' directory and is part of the vendor's local infrastructure.
- [DYNAMIC_EXECUTION]: The skill is designed to execute sequence-based logic defined in the 'activation_steps_prepend' and 'activation_steps_append' configuration fields. These fields allow for the dynamic execution of tasks defined in local override files (team or user levels).
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its data ingestion mechanisms. Ingestion points: Content is loaded from local files via glob patterns in 'agent.persistent_facts' and instructions are merged from project-level TOML configuration files. Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for content loaded from these external local files. Capability inventory: The skill possesses the capability to execute shell commands (via uv run) and sequence-based activation steps. Sanitization: Content loaded from local files or configuration overrides is interpolated directly into the agent context without sanitization or validation.
Audit Metadata