zentao-cli

Fail

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: HIGHCREDENTIALS_UNSAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill documents and encourages the storage of ZenTao credentials (passwords or tokens) in a hidden local configuration file at ~/.config/zentao/zentao.json. This persistent cache can be targeted for credential theft.
  • [CREDENTIALS_UNSAFE]: The documentation provides examples of passing raw passwords directly in shell commands (e.g., zentao login -p 123456), which leads to sensitive information being stored in the user's shell history.
  • [DATA_EXFILTRATION]: The skill instructs the AI to use the --yes flag for deletion operations to skip manual confirmation prompts. In an AI agent context, this allows for the silent and irreversible removal of remote ZenTao data without human oversight.
  • [EXTERNAL_DOWNLOADS]: The skill recommends global installation of the zentao-cli package using npm, bun, or pnpm without pinning to a specific version or providing integrity checks, making the environment vulnerable to supply-chain attacks.
  • [COMMAND_EXECUTION]: The skill relies on executing external shell commands to manage project data. While intended for its primary purpose, the combination of automated execution, credential access, and destructive capabilities increases the attack surface.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 1, 2026, 06:42 AM
Security Audit — agent-trust-hub — zentao-cli