Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
RAMBOXIE Claw Rpg
v0.1.1D&D-style RPG system for AI lobsters that auto-generates characters, tracks XP from conversations, manages leveling and prestige, and provides a web dashboard.
⭐ 0· 254·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The description promises an automated Node.js character system, web dashboard, and several scripts (init.mjs, easter.mjs, xp.mjs, setup-cron.mjs, etc.), but the skill package contains no code files or install spec. A functioning implementation would need those scripts, an npm project for the dashboard, and references — their absence is a major mismatch.
Instruction Scope
Runtime instructions explicitly tell the agent to read local files (SOUL.md, MEMORY.md), run node scripts after each reply (easter.mjs), update character.json/arena-history.json, and optionally execute child_process/execSync for XP sync. Those operations read and write local data and execute arbitrary local code; the SKILL.md also instructs integration into the reply pipeline, which can change outgoing messages. The instructions access files that are not declared elsewhere in the skill metadata and may contain sensitive content.
Install Mechanism
There is no install specification and no code files. Yet the instructions assume a Node/npm project and multiple scripts exist. That mismatch means the skill as distributed cannot operate without obtaining code from an external source — which is not provided or specified — increasing risk if the user downloads an unknown archive to satisfy the instructions.
Credentials
The skill declares no required environment variables or credentials (good), but instructs reading SOUL.md and MEMORY.md and writing character/config JSON files. Those files can contain private or sensitive data. It also references an optional Telegram config (chat id) and suggests using execSync from host code, which involves executing local commands that could be abused if inputs are not tightly controlled. The declared env/config surface is small, but actual instructions reach into local files and execute code.
Persistence & Privilege
always:false and no special privileges are requested. The SKILL.md recommends frequent invocation (after each reply or via cron), but that is a usage pattern rather than a platform-level persistence flag. Autonomous invocation remains possible (platform default).
What to consider before installing
Do not install or wire this skill into your agents unless you can obtain and review the missing code. The SKILL.md expects Node scripts, a dashboard, and local data files that are not included — which means you would have to download code from elsewhere. Before proceeding, ask the publisher for the source repository (git URL or release), review all scripts for network calls, filesystem writes, and child_process/exec usage, and confirm what SOUL.md and MEMORY.md contain (they may hold private data). If you want the feature, prefer a published package with a clear install path and a public source repo you can audit; avoid running unreviewed node scripts in production or attaching them to reply pipelines or heartbeat processes until reviewed.Like a lobster shell, security has layers — review code before you run it.
latestvk97dmzbx2zsf21myjc9x8js7bn82m7bw
License
MIT-0
Free to use, modify, and redistribute. No attribution required.