The Agent Skills Directory

[SAFE]: The skill provides legitimate operational guidance for triaging and managing inquiries (vimarshas) using internal domain-specific tools. No malicious behavior was identified.\n- [COMMAND_EXECUTION]: The skill utilizes a set of specialized MCP tools (nks_search, nks_look, nks_orient, nks_update, nks_arrow) for graph interaction. These tools are scoped to the internal knowledge management system and do not allow for arbitrary shell command execution or unauthorized system access.\n- [DATA_EXFILTRATION]: There are no indicators of sensitive data access or exfiltration. The skill does not access environment variables, credentials, or personal files, and it makes no external network calls.\n- [PROMPT_INJECTION]: The instructions are focused on operational logic and do not contain attempts to override the agent's system prompt or bypass safety protocols. The skill processes inquiry data which represents a standard indirect prompt injection surface. Ingestion points: Reads inquiry data via nks_search and nks_look. Boundary markers: None identified. Capability inventory: Graph modification via nks_update and nks_arrow. Sanitization: No specific content sanitization logic is described. This risk is inherent to the skill's primary function and is considered low given the restricted toolset.

inquiry