aliyun-oss-ossutil

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches official ossutil 2.0 binaries from Alibaba Cloud's content delivery network (gosspublic.alicdn.com) as documented in references/install.md.
  • [COMMAND_EXECUTION]: Invokes the ossutil CLI for storage management tasks (upload, download, sync, and bucket management). The utility script scripts/check_ossutil.py also executes 'ossutil --version' using the subprocess module to validate tool availability.
  • [PRIVILEGE_ESCALATION]: Installation steps in references/install.md use sudo for system-level operations, such as installing the unzip utility and moving the ossutil binary to /usr/local/bin. This is standard procedure for CLI tool installation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external storage (e.g., bucket names, object metadata) which represents a potential injection surface.
  • Ingestion points: Output from bucket listings and object metadata queries in SKILL.md.
  • Boundary markers: No specific delimiters are used to wrap external data in the provided workflows.
  • Capability inventory: File system write access to the output/ directory and network operations via the ossutil tool.
  • Sanitization: No explicit sanitization or filtering of object metadata is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 06:02 AM
Security Audit — agent-trust-hub — aliyun-oss-ossutil