iot-witness-verify

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the @claude-flow/plugin-iot-cognitum package from the npm registry using npx.
  • [REMOTE_CODE_EXECUTION]: Executes code from the downloaded npm package at runtime. The package namespace @claude-flow is consistent with the platform tools defined in the skill.
  • [COMMAND_EXECUTION]: Uses Bash to execute the verification plugin, passing user input directly as a command argument.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Evidence: 1. Ingestion points: device-id argument and command output processed by the agent. 2. Boundary markers: None. 3. Capability inventory: Bash and platform memory storage. 4. Sanitization: None.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 01:23 PM
Security Audit — agent-trust-hub — iot-witness-verify