bgp-registry-intel
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behavior or attack patterns were detected. The skill follows security best practices by implementing local checks that prevent sensitive internal IP addresses from being sent to public registries and provides robust instructional constraints to ensure accurate data interpretation.- [PROMPT_INJECTION]: The skill is designed to ingest data from external registries (RDAP, PeeringDB, RIPEstat), creating a potential surface for indirect prompt injection. This risk is effectively mitigated by specific instructions that require the agent to validate the source and age of the data and to follow specific logic tables for RPKI validation.
- Ingestion points: External API responses from
rpki-validator.ripe.net, RDAP, and PeeringDB. - Boundary markers: The skill requires all tool responses to include
sourceandcaveatsmetadata to ensure the agent maintains data provenance. - Capability inventory: Read-only network intelligence lookups via a local tool server.
- Sanitization: The skill contains explicit logic to refuse private and reserved IP addresses locally before any external request is made, preventing data exposure.
Audit Metadata