The Agent Skills Directory

[SAFE]: The skill consists entirely of behavioral instructions intended to transform the agent into a Socratic tutor. It does not request access to sensitive tools, perform network operations, or include any executable scripts.
[EXTERNAL_DOWNLOADS]: The README provides standard installation instructions using npx and git clone to retrieve the skill from its public repository. These are documented installation paths for the user and do not occur automatically or silently at runtime.
[SAFE]: The core logic focuses on restricting the agent's output (preventing direct answers) and managing a conversational workflow. This is a purely prompt-based interaction model with no evidence of prompt injection or bypass attempts.

socrates