aliexpress-search-by-image-reference-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 skill directs the agent to download and install the BrowserAct CLI and skills from an external GitHub repository (https://github.com/browser-act/skills) if they are not detected on the system.
- [COMMAND_EXECUTION]: The instructions explicitly command the agent to perform the installation "automatically without asking the user to confirm" and state "do not request separate approval for dependency installation." This is a direct attempt to bypass user oversight and agent safety guardrails regarding system modifications.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process data from external websites (AliExpress) and user-provided search queries, creating a vulnerability surface for malicious instructions embedded in remote content.
- Ingestion points: Marketplace target URLs, search queries, and seller profiles (SKILL.md, Instructions).
- Boundary markers: Not present; no delimiters or "ignore" instructions are defined for the scraped content.
- Capability inventory: The skill utilizes the browser-act tool which has network access and interactive browser capabilities.
- Sanitization: Not present; no validation or filtering is mentioned for the ingested data before it is normalized and returned.
Recommendations
- AI detected serious security threats
Audit Metadata