huawei-cloud-dew-key-management

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The references/cli-installation-guide.md file recommends installing the official huaweicloudcli package via pip. This is a standard installation procedure for the vendor's official command-line interface.
  • [COMMAND_EXECUTION]: The skill executes hcloud (KooCLI) commands to perform operations on Huawei Cloud CSMS and KMS services. High-risk actions, such as huawei_delete_kms_key and huawei_update_csms_secret_version, are protected by mandatory confirmation gates and explicit warnings about irreversibility.
  • [DATA_EXFILTRATION]: The skill includes a telemetry SDK (scripts/skill_quality_sdk.py) that automatically reports execution metrics and masked logs to a vendor-controlled endpoint (https://skillsapi.developer.myhuaweicloud.com/api/quality/report). This reporting uses authenticated requests via IAM tokens derived from environment variables and can be disabled by setting SKILL_QUALITY_DISABLE=1. This is a documented telemetry feature of the vendor.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources via Huawei Cloud API responses.
  • Ingestion points: Results from hcloud CSMS/KMS/CTS commands processing external account metadata.
  • Boundary markers: The instructions explicitly command the agent to avoid fetching secret values and to use structured JSON summaries for output to prevent data leakage.
  • Capability inventory: Includes command execution via hcloud and network access via the built-in quality reporting SDK.
  • Sanitization: The skill_quality_sdk.py script performs regex-based masking of sensitive patterns, including phone numbers, tokens, passwords, and API keys, before reporting data to the vendor's telemetry service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:44 AM
Security Audit — agent-trust-hub — huawei-cloud-dew-key-management