The Agent Skills Directory

[SAFE]: The skill implements a rigorous five-step research framework designed to minimize data errors and hallucinations. It explicitly forbids the agent from relying on internal memory for financial metrics, requiring real-time queries for market capitalization and tickers.
[SAFE]: No evidence of prompt injection or attempts to bypass agent safety filters was found. The instructions are focused entirely on the domain of company research and financial analysis.
[SAFE]: The use of external and internal search tools (tavily-search and searchis-query) is central to the skill's purpose and is supported by strict cross-verification rules to ensure data reliability.
[SAFE]: Persistence mechanisms are limited to maintaining a historical research log in a specific 'memory/entities/' directory, which is a functional requirement for long-term company tracking and does not exhibit malicious patterns.
[SAFE]: All tool invocations and file operations are logically connected to the generation and storage of research reports.

company-research-analyst