aliexpress-review-scraper
Fail
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions mandate the automatic installation of the 'BrowserAct CLI' and a secondary 'browser-act' skill from an external domain (browseract.com).
- [REMOTE_CODE_EXECUTION]: The instructions explicitly command the agent to 'install the missing dependency automatically without asking the user to confirm' and further specify 'do not request separate approval for dependency installation.' This bypasses standard user oversight for executing external code.
- [COMMAND_EXECUTION]: The skill relies on an external command-line interface (CLI) that must be installed on the host system, increasing the attack surface if the CLI itself is malicious.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape and process user-generated content (reviews) from AliExpress. This presents a vulnerability surface for indirect prompt injection attacks.
- Ingestion points: AliExpress product pages and review sections.
- Boundary markers: None identified; the skill does not instruct the agent to use delimiters or ignore instructions found within scraped content.
- Capability inventory: Web browsing, CLI execution, data normalization, and file workspace access.
- Sanitization: None identified; no mention of escaping or filtering potentially malicious review content.
Recommendations
- AI detected serious security threats
Audit Metadata