Skillv1.0.0

ClawScan security

Game Design Option Generation · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 21, 2026, 11:01 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only, self-contained game-design brainstorming skill whose declared purpose matches its files and requirements — no installs, credentials, or suspicious behavior are requested.
Guidance: This skill appears self-contained and safe: it only provides instructions for generating multiple game-design options and includes a small reference file. Before installing, consider what you will feed the skill — don't paste unreleased proprietary design docs or secrets into prompts. The skill does not require credentials or install anything, and it won't persistently modify the agent; still review outputs for confidentiality before sharing externally.

Purpose & Capability: okName/description ask for multiple design options; the skill contains only guidance and a small reference file and requests no binaries, env vars, or external services — all consistent with a design helper.
Instruction Scope: okSKILL.md instructs the agent to produce problem framing, multiple options, tradeoffs, and a recommendation. It only references the included family-conventions.md file for conventions and does not direct the agent to read unrelated files, access system state, or call external endpoints.
Install Mechanism: okNo install spec and no code files — the skill is instruction-only so nothing is written to disk or downloaded during install.
Credentials: okThe skill requires no environment variables, credentials, or config paths; requested access is proportional to a purely instructional/design skill.
Persistence & Privilege: okalways is false (default), and the skill does not request system persistence or modifications to other skills. Model invocation is allowed (platform default) but that is appropriate for an interactive design helper.