skills/aliyun/alibabacloud-aiops-skills/alibabacloud-cfw-vpc-peering-firewall-protect/Gen Agent Trust Hub
alibabacloud-cfw-vpc-peering-firewall-protect
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The script
scripts/validate-cli.shincludes instructions to download the Alibaba Cloud CLI installation script from the vendor's official CDN (aliyuncli.alicdn.com) and execute it via the shell. This is a standard vendor-provided installation method and originates from a trusted domain associated with the skill author. - [COMMAND_EXECUTION]: The skill makes extensive use of the
aliyunCLI to perform infrastructure operations and leveragespython3for internal logic, such as route table filtering, CIDR table extraction, and secure JSON assembly. - [INDIRECT_PROMPT_INJECTION]: The skill processes external infrastructure data including VPC route tables and firewall configurations, which represents an indirect injection surface. This is effectively mitigated through several layers of security:
- Ingestion Points: Data is fetched via API calls in
scripts/vpc-fw-switch.shandscripts/vpc-fw-lifecycle.sh. - Boundary Markers: The instructions enforce a strict three-phase workflow (Preview -> Confirm -> Execute) requiring explicit human approval for every write operation.
- Capability Inventory: Capabilities are limited to specific Cloud Firewall and VPC management APIs.
- Sanitization: All external data is sanitized using a
json_escapeutility inscripts/common.shbefore being interpolated into shell commands or JSON structures.
Audit Metadata