weibo-hot-search-scraper
Fail
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains instructions that explicitly direct the agent to bypass standard safety protocols and user consent by mandating the automatic installation of dependencies "without asking the user to confirm" and commanding the agent to "not request separate approval for dependency installation."
- [EXTERNAL_DOWNLOADS]: The instructions require the agent to fetch and install external software (BrowserAct CLI) from a third-party domain (browseract.com) that is not a recognized trusted vendor, which may lead to the execution of unvetted code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Weibo trending pages without security boundaries or sanitization instructions, creating an attack surface for malicious instructions embedded in the trending content.
- Ingestion points: Weibo hot search/trending pages specified in step 1 of the Instructions.
- Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are provided.
- Capability inventory: The skill utilizes live browser interaction via the
browser-acttool and export capabilities. - Sanitization: No data validation or sanitization protocols are defined for the retrieved content.
Recommendations
- AI detected serious security threats
Audit Metadata