walmart-category-listing-scraper
Fail
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The instructions in the 'BrowserAct Runtime' section direct the agent to automatically download and install the 'BrowserAct CLI' and associated skills from an external URL if they are not already present.
- [COMMAND_EXECUTION]: The skill explicitly commands the agent to suppress user notification and skip approval for dependency installation, stating 'do not request separate approval for dependency installation.' This bypasses human-in-the-loop security controls for system-level changes.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external Walmart listing pages, creating a vulnerability surface where malicious instructions embedded in product titles or metadata could influence agent behavior.
- Ingestion points: Walmart listing pages, product titles, and search results via the browser-act skill (SKILL.md).
- Boundary markers: None identified; instructions do not specify how to delimit external content from system instructions.
- Capability inventory: The skill has the capability to execute shell commands (CLI installation) and perform network operations via a browser tool.
- Sanitization: No evidence of input validation or sanitization for the scraped web content.
Recommendations
- AI detected serious security threats
Audit Metadata