design-md
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to clone an external repository (https://github.com/voltagent/awesome-design-md.git) to access the design specifications required for its operation.
- [COMMAND_EXECUTION]: The workflow defined in
SKILL.mddirects the agent to execute shell commands, specificallygit clone, to fetch the design library if it is not found in common local paths. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection because it processes data from an external source.
- Ingestion points: The agent reads brand-specific documentation from
<library>/design-md/<brand>/DESIGN.mdas specified in theSKILL.mdworkflow. - Boundary markers: The skill lacks explicit boundary markers or instructions to ignore embedded commands within the ingested design files.
- Capability inventory: The agent is instructed to "Follow it" and treat the design file as a "hard spec" for UI generation, potentially executing instructions embedded within the design documentation.
- Sanitization: There is no evidence of sanitization or validation of the content fetched from the remote repository before it is processed by the agent.
Audit Metadata