starworkSpawn

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a configuration generator for the starwork CLI, drafting blueprint and rule files. It does not perform network calls, access sensitive credentials, or execute arbitrary remote code.\n- [SAFE]: Instructions include clear safeguards against overwriting user files and specifically forbid modifications to system-critical directories such as .starwork/ or _系统/.\n- [PROMPT_INJECTION]: The skill functions as a surface for indirect prompt injection because it processes user-provided project descriptions into rules for downstream agents. This is an inherent architectural characteristic of template-generating skills.\n
  • Ingestion points: User project concepts and descriptions provided during the workflow.\n
  • Boundary markers: Not specified for the generated Markdown rule files.\n
  • Capability inventory: Drafting file structures and operational rules intended for the StarWork environment.\n
  • Sanitization: No specific filtering or validation of user input is directed before its inclusion in the generated files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:12 AM
Security Audit — agent-trust-hub — starworkSpawn