The Agent Skills Directory

[SAFE]: The skill implements secure defaults for memory training and data management. The memorization process strictly filters input to numeric digits, effectively minimizing the surface for indirect prompt injection. Data is stored in local JSON and Markdown files within the workspace, and the skill provides explicit warnings against using sensitive information like passwords or credit card numbers. No network activity, obfuscation, or remote code execution patterns were detected, and all conversion logic is deterministic based on user-verified mappings.

dominic-system