The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The content is purely informational and follows standard architectural practices.\n- [NO_CODE]: The skill consists exclusively of markdown documentation and conceptual code snippets for educational purposes. It does not include any executable scripts, configuration files, or external dependencies.\n- [PROMPT_INJECTION]: The documentation describes a pattern that processes untrusted user queries. This represents a theoretical architectural vulnerability surface (indirect prompt injection) that should be mitigated during actual implementation by using boundary markers or input sanitization.\n
Ingestion points: The user_query input parameter in the routing_workflow Python example (SKILL.md).\n
Boundary markers: None are present in the conceptual code snippets.\n
Capability inventory: None; the skill ships no executable code or tools.\n
Sanitization: Not addressed in the illustrative code examples.

routing