hmos-ability-insight-intent-generator

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides comprehensive instructions and templates for generating OpenHarmony intent decorators. It emphasizes adherence to official APIs and rigorous project configuration checks.
  • [DATA_EXPOSURE]: The skill instructs the agent to read local project configuration files (like build-profile.json5 and module.json5) and scan source code directories (/ets/pages/, /ets/database/, etc.). This data access is necessary for its primary purpose as a developer assistant and does not involve external exfiltration.
  • [COMMAND_EXECUTION]: The skill uses tools like Grep and Glob to explore the local project structure and content. This is a standard and safe practice for providing contextual code assistance.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted project code. However, it implements safeguards such as 'Bottom Line' rules (prohibiting fabrication and skipping checks) and a mandatory user confirmation flow before any file-writing operations occur.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 10:16 AM
Security Audit — agent-trust-hub — hmos-ability-insight-intent-generator