shop-app-research
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a bash helper script (
scripts/crawlora.sh) to interact with the Crawlora API. The script implements several security controls: it validates the format of theCRAWLORA_API_KEYto prevent shell injection, uses a temporary config file with restricted permissions (chmod 600) to pass the API key tocurlsecurely, and employs a strict regex-based whitelist to ensure only authorized API routes are accessed. - [EXTERNAL_DOWNLOADS]: The skill makes network requests to
api.crawlora.netto fetch shop and product data. These operations are restricted to the vendor's own infrastructure and use the HTTPS protocol for secure communication. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes external content, such as product reviews and merchant descriptions, which could contain adversarial instructions. However, the risk is minimal as the skill is limited to read-only
GETrequests and the data is retrieved through a structured API rather than raw HTML scraping.
Audit Metadata