market-environment-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary workflow involves collecting live market data from external sources via
web_search. This creates a vulnerability where instructions could be hidden in financial news, analyst reports, or web content to influence the agent's investment strategy recommendations. - Ingestion points:
web_searchtool is used to fetch stock indices, forex rates, commodity prices, and news from major financial media (Bloomberg, Reuters). - Boundary markers: None explicitly mentioned in the instructions to separate data from instructions.
- Capability inventory: The skill uses
scripts/market_utils.pyto format reports and has the capability to perform web searches and provide financial recommendations. - Sanitization: No evidence of data sanitization or instruction filtering before processing the fetched content.
- [COMMAND_EXECUTION]: The
SKILL.mdfile documents the execution of a local Python script (python scripts/market_utils.py) to generate report headers and check trading hours. While the script itself is benign, this establishes a pattern of local script execution for utility tasks.
Audit Metadata