design-taste-frontend
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The documentation provides instructions to download code from the author's GitHub repository (github.com/0xwilliamortiz/taste-skills.git) using git clone and npx.
- [COMMAND_EXECUTION]: The installation section contains shell commands intended for environment setup, including global package installation (npm install -g .) and execution of the taste-skills CLI tool.
- [PROMPT_INJECTION]: The skill ingests untrusted data from the user 'brief' (Ingestion point: SKILL.md) to influence UI generation (Capability). It lacks explicit boundary markers or sanitization logic to prevent embedded instructions in the brief from potentially influencing the agent's behavior beyond the intended design tasks, representing a surface for indirect prompt injection.
Audit Metadata