hedgefund-blackrock-portfolio
Warn
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill hardcodes an absolute path for executing a Python script:
/Users/engineer/.venv/bin/python3 .agents/skills/stocks-advisor/scripts/fundamentals.py. Referencing specific local user directories like/Users/engineeris a risky practice that targets specific environments and can be used to execute files outside of the intended skill sandbox or project structure.- [INDIRECT_PROMPT_INJECTION]: The skill ingests user financial data and passes it to internal logic and external scripts without defined safety boundaries. - Ingestion points: User profile intake (age, income, savings, goals, risk tolerance, and account type) in
SKILL.md. - Boundary markers: Absent. The skill does not instruct the agent to use delimiters or ignore embedded instructions within the user-provided data.
- Capability inventory: The skill utilizes command execution (Python scripts) and network tools (
web_fetch). - Sanitization: Absent. No input validation, filtering, or escaping is specified for the user-provided profile data before it is processed.
Audit Metadata