The Agent Skills Directory

[SAFE]: The skill implements standard financial modeling workflows using verified methodologies common in investment banking and equity research. Detailed instructions are provided for creating Excel files using Python's openpyxl library or Office JS, which are standard for this environment. No suspicious command execution or obfuscation was found in the provided files.
[EXTERNAL_DOWNLOADS]: The skill mentions external data sources like Tushare, AKShare, and SEC EDGAR. These are well-known, legitimate financial data providers used for institutional analysis. Referencing these sources for financial data is the primary purpose of the skill and does not pose a security risk.
[DATA_EXPOSURE]: The skill instructs the user on where to store configuration files and local repositories (e.g., ~/financial-services/), which is standard for local deployment. No patterns suggesting the exfiltration of sensitive files like .ssh or .aws credentials were observed.
[DYNAMIC_EXECUTION]: While the skill involves generating Excel formulas and Python code for financial modeling, this is restricted to the intended purpose of spreadsheet generation and data analysis. The guidelines emphasize 'Formulas Over Hardcodes' as a best practice for financial integrity rather than an evasion technique.

anthropic-finance-framework