Run a narrow compare between two or three backend profiles when architecture uncertainty is real. Use carefully, because it usually costs more tokens and time.

This skill appears coherent and low-risk, but review these before installing: (1) The skill expects to read/write project files (it targets docs/game-studio/backend-decision.md). Only grant file-system or repo write access in a workspace you trust, and require explicit confirmation before it modifies files. (2) The skill may recommend running builds or tests for best results — avoid giving it permission to run destructive commands and consider running any suggested builds/tests yourself in a sandbox. (3) Because it can modify documentation, require a human review/commit step (or use a branch/pull request) so you can inspect changes before they land. If the skill later requests network endpoints, credentials, or an installer, re-evaluate immediately — that would be a red flag.

Type: OpenClaw Skill Name: game-compare-backends Version: 1.0.0 The skill bundle is a set of markdown-based instructions designed to guide an AI agent in comparing game backend architectures. It contains no executable code, no network activity, and no evidence of malicious intent or prompt injection; its only functional requirement is updating a local documentation file (docs/game-studio/backend-decision.md) as specified in SKILL.md.