estimate-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_CONTEXT_INJECTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the
yfinancepackage from the public Python Package Index (PyPI). This is a well-known service and the package is standard for its stated purpose of financial data retrieval. - [COMMAND_EXECUTION]: The skill uses
subprocess.check_callto install dependencies viapipand executes a Python version check through a shell command. These are standard administrative tasks for environment setup. - [DYNAMIC_CONTEXT_INJECTION]: The
SKILL.mdfile uses the!command`` syntax to run a Python script at load time to verify the installation ofyfinance. This is used for configuration and does not access sensitive system files. - [DYNAMIC_EXECUTION]: The version check command uses the
exec()function to run a hardcoded Python snippet. This code is internal to the skill and does not process untrusted external data at that stage. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Yahoo Finance through the
yfinancelibrary as seen inSKILL.md. This ingestion point lacks explicit boundary markers or sanitization for external content, creating an attack surface where instructions embedded in financial data could influence the agent. However, the skill's capabilities are focused on data analysis and tabulation, minimizing the risk of exploitation.
Audit Metadata