finance-research-agent
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves financial data from several well-known external services to perform its research functions.
- Historical and current stock market data is fetched from Yahoo Finance via the yfinance library.
- Cryptocurrency price and market capitalization data is obtained from the CoinGecko API.
- Macroeconomic data series are pulled from the Federal Reserve Economic Data (FRED) API.
- News content is gathered from the Google News RSS feed and NewsAPI.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external news data, which creates a potential surface for instructions embedded in that data to influence agent behavior.
- Ingestion points: Untrusted data enters the agent context through news headlines and summaries fetched in scripts/finance_api.py (line 257).
- Boundary markers: There are no explicit markers or instructions used to delimit the external news content or warn the agent to ignore embedded commands.
- Capability inventory: The agent possesses the capability to perform network reads and write research reports to the file system using Jinja2 templates (scripts/finance_api.py, line 449).
- Sanitization: The skill lacks sanitization or filtering logic for the raw text retrieved from external RSS feeds before it is summarized or analyzed for sentiment.
Audit Metadata