harmony-empty-project

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes user-supplied strings such as project names and bundle IDs to generate directory structures and configuration files. This data is also used as arguments for shell commands. This creates a surface for indirect prompt injection where malicious input could potentially lead to path traversal or command injection if not properly sanitized by the underlying agent.\n
  • Ingestion points: User-provided project name and bundle ID in SKILL.md.\n
  • Boundary markers: Not present in the instructions.\n
  • Capability inventory: File writing across multiple directories and shell command execution (ohpm, hvigorw) in SKILL.md.\n
  • Sanitization: No specific sanitization or validation logic is defined within the skill for user-provided identifiers.\n- [COMMAND_EXECUTION]: The skill instructs the agent to execute development-related shell commands such as ohpm install and hvigorw for project setup and validation.\n
  • Evidence: Found in the '依赖管理与安装规则' and '构建前检查' sections of SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 10:36 AM