The Agent Skills Directory

[SAFE]: No malicious patterns, suspicious code, or security vulnerabilities were detected in this skill. The logic focuses exclusively on educational scaffolding and learning reinforcement.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data to generate simplified explanations and saved notes. 1. Ingestion points: User-provided concept, explanation attempt, and source materials in SKILL.md. 2. Boundary markers: Absent; user input is interpolated directly into the technique steps. 3. Capability inventory: Integration with connected notes tools (e.g., Scratchpad connector) for saving resulting content. 4. Sanitization: No explicit validation or instruction-filtering is applied to the user-provided data.

feynman-notes