skills/openclaw/skills/esp-idf-helper/Gen Agent Trust Hub

esp-idf-helper

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill's automation scripts (flash_with_progress.sh, monitor_auto_attach.sh, run_menuconfig.sh) dynamically load and execute environment setup files using the source command on paths provided via arguments (e.g., $IDF_DIR/export.sh). This is the standard mechanism for ESP-IDF but allows for the execution of arbitrary script content if the directory path is manipulated.
  • [COMMAND_EXECUTION] (LOW): The script usbipd_attach_serial.sh invokes powershell.exe from within the WSL environment to execute Windows-side commands for hardware management. This cross-OS execution is a functional requirement for WSL2 serial port access but expands the execution boundary.
  • [PROMPT_INJECTION] (LOW): The skill identifies as having an indirect prompt injection surface (Category 8) because it ingests and processes data from external tool outputs to drive automated retry logic.
  • Ingestion points: Command output from idf.py and usbipd list captured in scripts/flash_with_progress.sh and scripts/monitor_auto_attach.sh.
  • Boundary markers: Absent; scripts rely on simple grep pattern matching on command output.
  • Capability inventory: The skill can execute firmware flashing, serial monitoring, and cross-environment PowerShell commands.
  • Sanitization: Limited; the scripts use basic shell utilities like awk and tr to parse tool output without rigorous validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 09:31 AM