The Agent Skills Directory

[SAFE]: The skill's primary function is to guide the agent in facilitating learning through systematic questioning. The instructions focus on clarification, assumption-probing, and synthesis, which are safe instructional behaviors.
[SAFE]: Analysis of the referenced files (facilitation-guide.md and sample_scenario.md) shows they contain only educational content, structured scenarios, and facilitation tips. No executable scripts, network requests, or sensitive file access patterns are present.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input as the core 'starting position' for Socratic inquiry. Ingestion points: User-provided claims or beliefs are ingested into the prompt via the Starting Position field as defined in SKILL.md. Boundary markers: The skill uses a rigid output format and a specified Facilitation Flow to constrain the agent's output. Capability inventory: The skill has no code components and uses no external tools; its only capability is text generation within the conversation. Sanitization: The pedagogical framework requires the agent to analyze and question the input rather than execute instructions within it.

hum-socratic