The Agent Skills Directory

[SAFE]: The skill performs legitimate financial calculations and fetches data from well-known, trusted sources (yfinance, Yahoo Finance). The Python script scripts/capital_allocation.py uses only standard libraries and performs purely mathematical analysis on user-supplied or fetched financial data.
[COMMAND_EXECUTION]: The skill uses dynamic context injection in SKILL.md (e.g., !python3 -c ...) to check for dependencies and fetch live risk-free rates. These commands are benign, strictly limited to the intended functionality of the skill, and do not incorporate unvalidated user input into shell execution.
[EXTERNAL_DOWNLOADS]: The skill mentions auto-installing yfinance, numpy, and pandas. These are well-known, standard libraries in the Python data science ecosystem and are appropriate for the skill's purpose. The references to finance-data-providers:funda-data are for internal skill delegation.
[DATA_EXPOSURE]: The skill reads from a local directory .data/{TICKER}/{date}.json if it exists. This is described as a local caching mechanism to ensure data consistency within a reporting pipeline and does not access sensitive system files or user credentials.
[PROMPT_INJECTION]: The instructions are focused on financial modeling logic and methodology. There are no attempts to override agent safety filters or bypass platform constraints.

company-valuation