ziniao-account

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the ziniao-cli binary to perform administrative tasks including account creation, deletion, and authorization management. Destructive commands (e.g., account delete, account auth-clean) are appropriately guarded by a --yes flag requirement to prevent accidental execution.
  • [EXTERNAL_DOWNLOADS]: The skill specifies a requirement for the ziniao-cli binary. This tool is the vendor's official command-line interface for managing their SuperBrowser accounts and is consistent with the skill's stated purpose.
  • [PROMPT_INJECTION]: The skill processes external data such as account names, remarks, and tags which are supplied as arguments to CLI commands. This represents a potential indirect prompt injection surface; if these fields contain malicious instructions, they could influence the agent's behavior when the account data is subsequently listed or processed.
  • Ingestion points: account create, account update, and account tag-create commands (SKILL.md and references/ziniao-account-create.md).
  • Boundary markers: None explicitly defined in the prompts for the data fields.
  • Capability inventory: The skill can execute shell commands via ziniao-cli and access network APIs.
  • Sanitization: No specific sanitization or escaping of the user-provided names or remarks is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 03:31 AM