classic-text-adventure

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
  • Command Execution: The skill includes a diagnostic utility in scripts/smoke_test.py that uses the subprocess module. This is used exclusively to run the bundled runner.py script to verify the game engine's integrity and environment compatibility during setup or diagnostic requests.
  • Unsafe Deserialization: The bundled game engine library contains a legacy resume method in scripts/runtime/adventure/game.py that uses the pickle module. However, the skill's primary runtime implementation in scripts/runner.py and scripts/runtime_adapter.py intentionally bypasses this path. It employs a safer JSON-based transcript replay mechanism to restore game state, which mitigates the standard risks associated with pickle-based persistence.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 11:55 PM
Security Audit — agent-trust-hub — classic-text-adventure