awesome-design-html
Warn
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct users to install the skill by cloning a repository from GitHub (
github.com/yzfly/awesome-design-html) or downloading ZIP files from its releases page. It also makes external requests to Google Fonts (fonts.googleapis.com) to load typography. - [COMMAND_EXECUTION]: The installation and troubleshooting sections involve executing shell commands including
git clone,git pull,rm -rf, andlsto manage the skill's file structure and updates. - [PROMPT_INJECTION]: The skill implements a workflow for indirect prompt injection by instructing the agent to dynamically load and process local HTML files based on user-supplied 'brand' names.
- Ingestion points: The agent is instructed to read from
assets/web/design.[brand].htmlandassets/ios/design.[brand]-ios.html(SKILL.md). - Boundary markers: No delimiters or safety instructions are provided to ensure the agent ignores any malicious instructions that might be embedded within the design reference HTML files.
- Capability inventory: The agent uses file system access tools to retrieve the design tokens and structure from the reference library.
- Sanitization: The instructions lack validation or sanitization for the 'brand' input, which could potentially be exploited for path traversal to read unintended files if not handled correctly by the underlying agent platform.
Audit Metadata