huawei-cloud-waf-aad-rule-management

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download the hcloud CLI installation script from https://cn-north-4-hcli-cloud.s3.cn-north-4.myhuaweicloud.com/install.sh and execute it via bash. This is a vendor-provided resource for the official Huawei Cloud command-line tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from cloud APIs (such as policy names, rule configurations, and domain names), which presents a surface for indirect prompt injection attacks.
  • Ingestion points: Data returned from hcloud WAF ListCompositeHosts, ListPolicy, ListCustomRules, and AAD ListInstance (defined in references/verification-method.md).
  • Boundary markers: None identified in the prompt templates or instructions to delimit tool output.
  • Capability inventory: The skill can perform powerful write operations including creating and deleting firewall rules using hcloud WAF BatchCreateRule and DeleteRule commands (defined in SKILL.md).
  • Sanitization: No explicit sanitization or validation of the ingested cloud resource names or configurations is mentioned in the skill logic.
  • [SAFE]: The skill includes a local SDK script (scripts/skill_quality_sdk.py) that reports execution quality metrics to a vendor-owned endpoint (skillsapi.developer.myhuaweicloud.com). The SDK implements data masking for sensitive information such as phone numbers, secrets, and authentication tokens before transmission.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:44 AM
Security Audit — agent-trust-hub — huawei-cloud-waf-aad-rule-management