The Agent Skills Directory

[SAFE]: The skill operates entirely within the local environment, reading project documentation and codebase to assist the user in business positioning.
[SAFE]: A robust confirmation mechanism is required before the agent writes the .agents/memory/icp.md file, ensuring user oversight of the generated content.
[NO_CODE]: The skill consists of instructions and metadata without any attached scripts, executables, or external package dependencies, significantly reducing the attack surface.

icp