yfinance-data
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill automatically installs the
yfinancePython package using the system's pip installer if the library is not already present. This is a standard dependency management practice for agent skills. - [COMMAND_EXECUTION]: The skill uses
subprocess.check_callto execute the pip installation command. The command uses static arguments and is used only for environment preparation. - [DYNAMIC_CONTEXT_INJECTION]: The
SKILL.mdfile utilizes the dynamic context injection syntax (!python3 -c ...) to check if theyfinancelibrary is installed and verify its version. The shell command is benign, static, and does not incorporate user-supplied input. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Yahoo Finance, including news articles, company summaries, and analyst reports. This external data represents an indirect prompt injection surface.
- Ingestion points: Financial data and news summaries retrieved via
yfinance.Tickermethods such as.info,.history(), and.news. - Boundary markers: No specific boundary markers or instructions to treat external data as untrusted are defined in the instructions.
- Capability inventory: The skill allows Python code execution and includes a subprocess call capability for package management.
- Sanitization: There is no evidence of filtering or sanitization applied to the external market data before it is provided to the agent for analysis.
Audit Metadata