netbox-review-datamodel
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to audit and review live NetBox data models, which involves ingesting untrusted data from an external MCP server. This creates a surface for indirect prompt injection where malicious content within the NetBox instance (such as object descriptions or custom field values) could attempt to influence the agent's behavior.
- Ingestion points: Data model object names, descriptions, and custom field values retrieved from a NetBox MCP server as specified in the 'Retrieval Sources' section of SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or warnings to treat audited content as untrusted data.
- Capability inventory: The skill is intended for use with a NetBox MCP server to 'inspect the current data model, count objects, find inconsistencies.' The agent's broader capabilities (file system or network access) could be targeted via this ingestion surface.
- Sanitization: There are no instructions provided for sanitizing or validating the input data before processing it for the audit review.
Audit Metadata