The Agent Skills Directory

[SAFE]: The skill uses official and well-known data sources, including the SEC EDGAR system and the FRED macro database. All external communications are conducted through standard, reputable libraries (yfinance, requests) to established domains.
[DYNAMIC_EXECUTION]: A benign use of __import__ is found in scripts/common/utils.py within the require_dependency function. This is used only to verify that necessary libraries are installed before a script runs, with module names provided as hardcoded strings.
[INDIRECT_PROMPT_INJECTION]: Although the skill ingests external data from financial APIs and filings, it does not possess exploitable capabilities like arbitrary command execution or file writes, which significantly limits the risk of indirect prompt injection.

findata-toolkit-us