bolt-iot

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the official Membrane CLI tool (@membranehq/cli) from the public npm registry to facilitate communication with the IoT platform.
  • [COMMAND_EXECUTION]: Utilizes the membrane CLI to execute shell commands for device management, including reading sensors, writing to serial ports, and controlling GPIO pins.
  • [PROMPT_INJECTION]: The skill acts as a bridge for data from external IoT devices, which presents a surface for indirect prompt injection if those devices return untrusted content.
  • Ingestion points: Output from API requests and device actions in the SKILL.md file.
  • Boundary markers: No explicit delimiters are used to separate external IoT data from the agent's internal instructions.
  • Capability inventory: Includes shell command execution and device state modification across all integrated scripts.
  • Sanitization: Responses from the Bolt IoT API are processed by the agent without specific sanitization or filtering logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 09:33 PM