The Agent Skills Directory

[SAFE]: No malicious patterns, exfiltration, or unauthorized code execution were detected. The skill is focused on local data persistence and organization.
[COMMAND_EXECUTION]: The skill instructs the agent to use the rg (ripgrep) CLI tool to perform searches within the ./memory/ directory. These commands are scoped to the local environment and are used for keyword and tag-based retrieval of saved notes.
[PROMPT_INJECTION]: The skill's primary function—storing and recalling user-influenced content—creates a surface for indirect prompt injection. If an agent saves and later recalls malicious instructions provided during a previous session, it could influence future behavior.
Ingestion points: Files stored within the ./memory/ directory (referenced throughout SKILL.md).
Boundary markers: None identified in the provided templates for memory files.
Capability inventory: Local file read/write operations and ripgrep search functionality (referenced in SKILL.md).
Sanitization: The skill does not provide explicit mechanisms for sanitizing or escaping content before it is re-injected into the agent's context during retrieval.

memory-bank