nginx-to-higress-migration

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: All identified behaviors, such as reading Ingress configurations and deploying components via Helm, align with the skill's primary purpose and use official vendor-controlled infrastructure.
  • [EXTERNAL_DOWNLOADS]: Fetches infrastructure components and configuration from official Higress project domains and Alibaba Cloud container registries.
  • [COMMAND_EXECUTION]: Uses standard administrative tools (kubectl, helm, docker) to perform cluster discovery, installation, and plugin image management.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it processes untrusted data to generate code. 1. Ingestion points: Reads cluster-wide Ingress resources and Nginx ConfigMaps via kubectl (SKILL.md, analyze-ingress.sh). 2. Boundary markers: No explicit markers or instructions to ignore embedded instructions are present in the processing logic. 3. Capability inventory: Includes subprocess calls for kubectl apply, docker push, and go build across the migration scripts. 4. Sanitization: No content sanitization or validation of Ingress annotations is performed before AI processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 03:18 AM
Security Audit — agent-trust-hub — nginx-to-higress-migration