skills/httprunner/skills/harmony-hdc/Gen Agent Trust Hub

harmony-hdc

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides templates for executing arbitrary shell commands on a target HarmonyOS device via the hdc shell <command> syntax. This grants the agent significant control over the connected device's operating system environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection through the processing of device-generated data.
  • Ingestion points: The agent is instructed to capture screenshots (snapshot_display) and dump UI layout trees (uitest dumpLayout -p /data/local/tmp/layout.json) which it then retrieves via hdc file recv to process.
  • Boundary markers: The instructions do not define boundary markers or explicit warnings for the agent to ignore instructions embedded within the UI text or layout metadata.
  • Capability inventory: The skill includes high-privilege capabilities including raw shell command execution (hdc shell), file transmission (hdc file send), and application installation (hdc install).
  • Sanitization: There is no evidence of sanitization or validation logic for the content retrieved from the device before it is interpreted by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 11:30 AM