Skills
skills/parcadei/continuous-claude-v3/connectedness/Gen Agent Trust Hub

connectedness

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions focus on mathematical strategies for topology (connectedness) and do not contain any prompt injection or bypass attempts.
  • [COMMAND_EXECUTION]: The skill defines commands using uv run to execute local Python scripts (scripts/z3_solve.py and scripts/sympy_compute.py). These are standard operations for mathematical verification and do not involve untrusted remote sources or arbitrary command injection.
  • [DATA_EXPOSURE]: No sensitive file paths, credentials, or network exfiltration patterns were identified. The skill operates on mathematical expressions and local logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 01:57 PM