zentao-shared

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to monitor the JSON output of the zentao-cli tool for a specific _notice.update field. If detected, the agent is prompted to suggest that the user execute update commands. This mechanism relies on external tool output to trigger agent behavior.
  • Ingestion points: Tool output from zentao-cli executions.
  • Boundary markers: None specified for the processing of tool output.
  • Capability inventory: The skill allows configuration and execution of global package updates (npm update) and CLI authentication.
  • Sanitization: No sanitization or validation of the update notification content is described before it influences the agent's response.
  • [COMMAND_EXECUTION]: The skill provides explicit instructions for the agent to execute shell commands to configure the tool (zentao-cli config set), authenticate users (zentao-cli auth login), and perform global updates (npm update -g zentao-cli).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 01:26 AM
Security Audit — agent-trust-hub — zentao-shared