The Agent Skills Directory

[SAFE]: The skill's primary function is to manage local knowledge records and user preferences without performing any dangerous network or system operations.
[DATA_EXPOSURE]: The provided management script scripts/profile_store.py includes an explicit security filter (SENSITIVE_PARTS) that prevents the storage of fields containing sensitive keywords such as 'token', 'secret', 'password', 'credential', or 'api_key'.
[COMMAND_EXECUTION]: The skill provides Python utility scripts (profile_store.py and validate_skill.py) for maintaining user configurations and validating the integrity of the skill source. These scripts use safe practices, such as yaml.safe_load and atomic file writes with restricted permissions (0o600).
[INDIRECT_PROMPT_INJECTION]: While the skill can ingest content from external URLs to generate knowledge notes, its workflow is designed to prevent fact hallucination by fetching actual content rather than relying on titles, and it stores this data in a structured, local-only format.

lov-memory-add