alibabacloud-remote-skills-connector

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions include a command to fetch the official Alibaba Cloud CLI installation script from the vendor's CDN (aliyuncli.alicdn.com). This is documented as a prerequisite for using certain credential sources.
  • [REMOTE_CODE_EXECUTION]: The skill provides a command to pipe a remote setup script directly into bash (curl -fsSL https://aliyuncli.alicdn.com/setup.sh | bash). While this is a high-risk execution pattern, it targets the official vendor's domain and is used for platform setup.
  • [COMMAND_EXECUTION]: The bundled Python scripts (scripts/agenthub_runtime/session.py and scripts/a2a_proxy/references/client_detect.py) use the ps utility to inspect parent processes. This is used legitimately to identify the hosting AI agent environment (e.g., Claude Code, Qwen Code) and ensure session isolation.
  • [PROMPT_INJECTION]: Automated scans flagged references/ram-policies.md for concealment. Review of the content shows this is a false positive; the file provides technical guidance on Alibaba Cloud RAM (Resource Access Management) policies, explaining that a specific API is allowed by default and how to troubleshoot permission errors without reducing user security.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 06:19 AM
Security Audit — agent-trust-hub — alibabacloud-remote-skills-connector