The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses shell commands to interact with the NetBox API. This includes using curl for REST operations (GET, POST, PATCH, DELETE) and python (via -m json.tool or -c scripts) for parsing and extracting data from JSON responses. These operations are directed at the infrastructure defined by the user's $NETBOX_URL environment variable.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface in its workflow for automated remediation.
Ingestion points: Validation findings and results are ingested from the NetBox API (e.g., /api/plugins/validation/findings/), as documented in SKILL.md and references/api-patterns.md.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the documentation for handling findings.
Capability inventory: The skill grants the agent the ability to execute network requests via curl to modify NetBox data and perform validation runs across all referenced files.
Sanitization: There is no evidence of sanitization or filtering applied to the validation findings before they are analyzed by the agent.

netbox-validation