permission-manager

Fail

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: HIGHCOMMAND_EXECUTIONDATA_EXFILTRATIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill enables the agent to alter its own security policy by adding or removing commands from authorized lists within the configuration.
  • [DATA_EXFILTRATION]: The skill accesses and summarizes the contents of the ~/.config/opencode/opencode.json file, which is a sensitive system configuration file governing the agent's runtime permissions.
  • [NO_CODE]: The skill contains no executable scripts or binaries, relying instead on natural language instructions to direct the agent's interaction with its own configuration files.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 14, 2026, 08:33 AM
Security Audit — agent-trust-hub — permission-manager