skills/lvlup-sw/exarchos/synthesis/Gen Agent Trust Hub

synthesis

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill uses shell commands only for local project validation (npm scripts) and Git operations, which are standard for the synthesis phase.
  • [DATA_EXPOSURE]: No hardcoded credentials or unauthorized access to sensitive files were detected; it uses project-specific state files for workflow management.
  • [PROMPT_INJECTION]: The instructions are operational and do not attempt to override system safety protocols or extract system prompts.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes branch names and PR feedback, these inputs are used within a validated development workflow without exposing dangerous execution surfaces.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 04:36 AM