The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation instructs the agent to execute the ov add-memory command to persist conversation data. This is a local command execution associated with the OpenViking toolset.
[PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by capturing untrusted conversation data and storing it in a persistent database for future retrieval. * Ingestion points: Conversation role and content data passed to the ov CLI. * Boundary markers: None specified in the usage examples. * Capability inventory: Local system command execution via the ov tool. * Sanitization: No explicit input validation or sanitization is described in the skill.

adding-memory