amazon-movers-shakers
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/scrape_amazon.pyutilizes thesubprocess.check_callfunction to programmatically execute system commands. Specifically, it invokes the Python package manager (pip) to install dependencies. - [EXTERNAL_DOWNLOADS]: Both
scripts/setup.shandscripts/scrape_amazon.pyare designed to download and install external Python packages (requests,beautifulsoup4) from public registries during setup or runtime. - [INDIRECT_PROMPT_INJECTION]: The skill functions as a data ingestion surface by retrieving product titles, rankings, and trends from external sources.
- Ingestion points: Data is fetched from
amazon.comdomains via web scraping and fromapi.sellersprite.comvia API calls, both of which are attacker-controllable environments (e.g., via malicious product listings). - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to disregard potential prompts embedded within the scraped product data.
- Capability inventory: The skill possesses network access capabilities and the ability to execute shell commands for environment setup.
- Sanitization: The script performs basic parsing but does not explicitly sanitize or filter the content of product titles or descriptions for potential injection patterns before passing them to the agent context.
Audit Metadata