store-teardown
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as intended for store design analysis, using ScraperAPI and Google Gemini API through authorized environment variables. No malicious behaviors such as credential exfiltration or unauthorized file access were detected.
- [PROMPT_INJECTION]: The skill processes untrusted product titles and shop names from external e-commerce sites, which represents a surface for indirect prompt injection. Ingestion points: Product titles and metadata fetched from Taobao, Tmall, and Instagram in Phase 1 and stored in items.json. Boundary markers: No specific delimiters or safety instructions are used in the Phase 2 analysis prompts to separate data from instructions. Capability inventory: Performs network requests to external APIs (ScraperAPI, Google Search, Gemini API) and file writes to /tmp/. Sanitization: Ingested text is included directly in prompts without escaping or validation.
Audit Metadata