investment-research
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external financial websites and user-supplied arguments to generate reports and commands.
- Ingestion points: External data sources (Macrotrends, StockAnalysis, AASTOCKS, etc.) and the
$ARGUMENTSvariable inSKILL.md. - Boundary markers: No explicit delimiters are used to isolate untrusted data within the prompt or shell commands.
- Capability inventory: Access to
bashfor running repository tools, file system access to write reports, and web search capabilities. - Sanitization: While the skill mandates numeric validation via
tools/financial_rigor.py, it lacks comprehensive sanitization for string data used in shell commands. - [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (e.g.,
python3 tools/financial_rigor.py verify-market-cap --price {股价} ...) where arguments are directly populated from data scraped from the web. - [DYNAMIC_EXECUTION]: Shell command strings are dynamically assembled at runtime based on variables fetched from external sources, which could lead to command injection if the source data contains shell metacharacters.
Audit Metadata