Skills
skills/huaweicloud/huaweicloud-skills/huawei-cloud-flexus-l-server-openclaw-deployment/Gen Agent Trust Hub

huawei-cloud-flexus-l-server-openclaw-deployment

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches installation scripts from documentation-samples.obs.cn-north-4.myhuaweicloud.com. This domain is confirmed to be the official Huawei Cloud Object Storage Service used for sample deployments.
  • [REMOTE_CODE_EXECUTION]: Shell scripts retrieved from the vendor's official domain are executed locally and on remote targets via piped-to-bash commands. This is a functional requirement of the deployment process to automate environment setup.
  • [COMMAND_EXECUTION]: Local system tasks and service restarts are performed using subprocess.run with shell=True. These operations are limited to the scope of deploying and managing the OpenClaw instance.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection.
  • Ingestion points: Configuration data is ingested via the model_params parameter in scripts/models.py and the channel_list parameter in scripts/channels.py.
  • Boundary markers: There are no explicit boundary markers separating configuration data from the shell command templates.
  • Capability inventory: The skill is capable of local command execution via subprocess.run and remote execution via the Cloud Operations Center (COC) API as defined in scripts/lib.py.
  • Sanitization: The normalize_json_string function provides basic JSON structural validation but does not sanitize inputs against shell metacharacters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 09:10 AM
Security Audit — agent-trust-hub — huawei-cloud-flexus-l-server-openclaw-deployment