The Agent Skills Directory

[SAFE]: The skill provides a secure and structured approach to migrating hardware definitions from NetBox to Infrahub. The logic is encapsulated in Python scripts that use standard parsing libraries and follow best practices for data handling.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions to download hardware definitions from the official NetBox community repository on GitHub (github.com/netbox-community/devicetype-library). This is a trusted and well-known source within the network automation ecosystem and is essential for the skill's functionality.
[INDIRECT_PROMPT_INJECTION]: The skill processes external YAML data from NetBox libraries, which serves as an entry point for untrusted content. However, the risk is mitigated by the skill's design:
Ingestion points: scripts/netbox_to_infrahub_templates.py reads NetBox YAML files.
Boundary markers: The agent is guided by specific rules and mapping profiles that define exactly how data is interpreted.
Capability inventory: The skill allows Bash for running the converter and infrahubctl, but these tools are used to process data into static YAML artifacts rather than executing it.
Sanitization: The converter uses yaml.safe_load to prevent arbitrary code execution during parsing. Given that this ingestion is the primary purpose of the skill, the risk is considered negligible.
[DYNAMIC_CONTEXT_INJECTION]: The SKILL.md file contains dynamic context injection placeholders using the !`find` command. These are used to display relevant project files (schemas and mapping profiles) to the agent upon loading. These commands are benign, strictly limited to project navigation, and do not involve network exfiltration or sensitive file access.

infrahub-converting-netbox-device-types