investment-research
Warn
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run shell commands (e.g.,
python3 tools/financial_rigor.py) using arguments populated with data retrieved from external financial websites (like{股价}or{总股本}). Without explicit sanitization of these variables, this pattern allows for potential command injection if a data source or user input is maliciously crafted. - [EXTERNAL_DOWNLOADS]: The research workflow involves fetching data from well-known financial services, including East Money, Macrotrends, AAStocks, and CNInfo.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. • Ingestion points: External financial data from multiple sources (e.g., Macrotrends, East Money) and user-provided
$ARGUMENTS. • Boundary markers: None identified; external content is processed without delimiters. • Capability inventory: Execution of shell commands and writing to the local filesystem (~/). • Sanitization: No validation or escaping is performed on external strings before they are used in commands or file operations.
Audit Metadata