iot-protocols

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from external IoT devices.
  • Ingestion points: Data enters the system via commands like openclaw iot get and API endpoints like POST /api/iot/send which handle arbitrary payloads from sensors and gateways (SKILL.md).
  • Boundary markers: There are no specified delimiters or instructions to treat device-provided data as untrusted text within the provided usage patterns.
  • Capability inventory: The skill possesses the ability to perform network operations, execute CLI subcommands, and pipe data to other services or databases.
  • Sanitization: The instructions do not mention sanitizing, escaping, or validating the content of data payloads (e.g., from CoAP or LoRaWAN) before they are used in subsequent operations or logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 05:44 PM