quick-design
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided arguments and content from existing design files to generate new specifications. This creates an attack surface where instructions embedded in analyzed data could influence the agent during the document generation process. Ingestion points:
argument-hintand files read fromdesign/gdd/andassets/data/. Boundary markers: Absent. Capability inventory: Write, Edit, and restricted Bash execution. Sanitization: Absent. - [DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context injection (the
!syntax) to execute a local configuration helper script at load time. The command is hardcoded to a local project script and is used for non-malicious environment setup. - [COMMAND_EXECUTION]: The skill is configured to use the
Bashtool, but it reduces risk by restricting usage to a specific local script path (yaml-helper.sh) within the project's hooks directory, preventing the execution of arbitrary shell commands.
Audit Metadata