skills/mrexodia/ida-pro-mcp/idapython/Gen Agent Trust Hub

idapython

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (HIGH): The skill establishes an attack surface for Indirect Prompt Injection. It documents functions to ingest untrusted data from binary files (e.g., strings and names) while providing high-privilege execution capabilities. Malicious binaries can contain crafted data that, when interpreted by the agent, override its instructions.\n
  • Ingestion points: idautils.Strings() (docs/idautils.md), ida_name.get_name() (docs/ida_name.md), ida_bytes.get_bytes() (docs/ida_bytes.md).\n
  • Boundary markers: None.\n
  • Capability inventory: ida_dbg.start_process(), ida_expr.exec_idc_script(), ida_bytes.patch_bytes(), ida_loader.load_plugin().\n
  • Sanitization: None.\n- [COMMAND_EXECUTION] (HIGH): The ida_dbg module (docs/ida_dbg.md) exposes functions to start and attach to processes (start_process, attach_process), enabling the agent to execute arbitrary code on the host system via the debugger.\n- [REMOTE_CODE_EXECUTION] (HIGH): The ida_expr module (docs/ida_expr.md) allows the agent to compile and execute IDC scripts (exec_idc_script, compile_idc_text), which facilitates dynamic code execution that could be exploited if the agent processes untrusted inputs.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 09:47 PM