aiox-data-engineer
Warn
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The Activation Protocol requires the agent to execute a Node.js script located at
.aiox-core/development/scripts/generate-greeting.js. This represents dynamic command execution on the host environment during the initialization of the skill. - [PROMPT_INJECTION]: The skill defines a
*yolocommand which allows the agent to cycle through permission modes, specifically 'auto' and 'explore'. These modes are designed to reduce or remove user-in-the-loop validation for agent actions, which can be exploited to perform unauthorized operations if the agent's context is compromised. - [PROMPT_INJECTION]: The activation protocol instructs the agent to 'Adopt this agent persona and command system' based on external files in
.aiox-core/and.codex/. Loading configuration and instructions from these paths can introduce indirect prompt injection if the files are modified by an attacker to include malicious instructions.
Audit Metadata