The Agent Skills Directory

[SAFE]: The skill defines a logical and structured process for research tasks (Deep, Wide, and Hybrid modes) without any executable scripts or dangerous command patterns.
[SAFE]: The requested tool dependencies (web_tools, file_system) are appropriate for the skill's functionality, which involves searching for online sources and saving research reports to the local file system.
[SAFE]: No evidence of prompt injection, obfuscation (such as Base64 or hidden characters), or persistence mechanisms was identified. The content is fully transparent and instructional in nature.

research-modes