wuji-cli-support

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the wuji CLI to perform diagnostic tasks such as device discovery (wuji devices), connectivity tests (wuji ping), health checks (wuji doctor), and log exportation (wuji logs export). These are standard operations for a device support utility.
  • [EXTERNAL_DOWNLOADS]: The skill references official troubleshooting documentation at docs.wuji.tech. This is a legitimate vendor resource used to match observed symptoms with known issues.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes local system logs and external documentation which are potentially untrusted sources. However, it includes a mitigation instruction to treat log text as evidence rather than executable instructions.
  • Ingestion points: Local device logs, manifest.json, doctor_diagnosis.json, and official documentation pages from docs.wuji.tech.
  • Boundary markers: The skill relies on natural language instructions to distinguish data from instructions rather than technical delimiters.
  • Capability inventory: The skill can execute wuji CLI commands and read local files to construct a report.
  • Sanitization: The instructions explicitly require privacy redaction of logs and credentials before inclusion in the final report.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 07:03 AM