mt5-log-reader
Pass
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill is granted access to the
Bashtool, which is used for calculating current dates and constructing file paths. While the defined use cases are benign, providing a shell environment increases the overall attack surface.\n- [DATA_EXFILTRATION] (LOW): The skill accesses MT5 log files that may contain sensitive data, such as account identifiers, trading activity, and proprietary logic results. Although there is no mechanism for external network exfiltration in the skill, the exposure of this data to the LLM context is a minor risk factor.\n- [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection as it ingests untrusted data from log files generated by external trading software.\n - Ingestion points: Runtime logs and compilation logs ($MQL5_ROOT/.../*.log).\n
- Boundary markers: Absent. The skill does not implement delimiters or instructions to ignore embedded commands within the log data.\n
- Capability inventory: The skill has access to
Read,Grep, andBashtools.\n - Sanitization: Absent. Log contents are processed and searched directly without escaping or validation, potentially allowing malicious log entries to influence the agent's logic.
Audit Metadata