wechat-fastest-growing
Fail
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: HIGHDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The script
scripts/fetch_growth_rank.pyprogrammatically opens and reads sensitive system files, including~/.bashrc,~/.zshrc,~/.bash_profile, and~/.profile, to extract API keys. This data is subsequently sent to the vendor's API endpoint. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from a remote API which is processed and rendered by the agent.
- Ingestion points: Article titles, summaries, and content are fetched via
requests.getinscripts/fetch_growth_rank.py. - Boundary markers: The skill instructions do not define delimiters or provide instructions to the agent to ignore potential commands embedded in the external data.
- Capability inventory: The skill possesses file system read access and network communication capabilities.
- Sanitization: No sanitization, escaping, or validation of the remote article data is performed before it is output to the user and the agent.
- [COMMAND_EXECUTION]: The skill relies on the execution of a local Python script
scripts/fetch_growth_rank.pyto perform its primary data retrieval and formatting tasks.
Recommendations
- AI detected serious security threats
Audit Metadata