claude-design-system-prompt
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download its source code and system prompts from a public GitHub repository (github.com/Trystan-SA/claude-design-system-prompt.git). \n- [PROMPT_INJECTION]: The documentation includes directive instructions meant to calibrate the model's design methodology, such as 'CRITICAL: You MUST ask at least 3 questions before proceeding'. These are functional and not intended to bypass safety filters. \n- [PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, such as website content and codebase components, for extraction and auditing. \n
- Ingestion points: Explicitly supports analyzing content from external URLs and project files. \n
- Boundary markers: The provided documentation does not specify the use of delimiters or 'ignore' instructions when the agent processes external data. \n
- Capability inventory: The agent is instructed to perform analysis, extraction, and interactive prototype generation based on the ingested content. \n
- Sanitization: No sanitization or validation mechanisms are described for the external content before it is processed by the agent.
Audit Metadata