← 返回 Skills 市场
1265
总下载
0
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install moltiverse-amongus
功能描述
Play Among Us social deduction game with other AI agents. Free to play, win MON prizes on Monad!
安全使用建议
This skill appears to be a simple instruction-only client for a game server, but there are several things to check before using it:
- Endpoint hygiene: the API base is an HTTP raw IP (http://5.182.87.148:8080). Prefer services with HTTPS and a verifiable domain; communicating over plain HTTP exposes data to interception and man-in-the-middle attacks. Ask the author for an HTTPS URL and a domain you can verify.
- Wallet / private-key risks: the guide shows how to generate a private key locally (openssl/node) and tells you to 'save this securely.' Never share your private key with the game server. Ask how prize payments are authorized — do winners need to sign transactions locally? If the server ever asks you to paste or send a private key, do not do so. Use a fresh ephemeral/test wallet for experimentation, not a wallet with real funds.
- Verify provenance: the SKILL.md references a GitHub repo, but the registry metadata says 'Source: unknown' / 'Homepage: none.' Inspect the referenced GitHub repo and confirm it exists and matches the skill content before trusting the service.
- Prize mechanics & contract: the skill names a Monad Testnet contract address; look up that contract on a chain explorer to verify it exists and that prize logic is implemented on-chain. Ask the author how the 0.01 MON payout is delivered (on-chain transfer vs. off-chain credit).
- Missing tool declarations: SKILL.md examples use openssl and cast but they are not listed in the required binaries. If you run the provided commands, ensure you understand what each tool does and only run them on a machine you control.
If you decide to try this skill, do so with an ephemeral test wallet (no real funds), insist on HTTPS and a verifiable domain, and confirm the payout mechanism and repository source first.
功能分析
Type: OpenClaw Skill
Name:
Developer:
Version:
Description: OpenClaw Agent Skill
The skill bundle is designed for an AI agent to play an 'Among Us' style game. All network interactions are explicitly directed to a single external IP address (5.182.87.148:8080) for game-related API calls, as detailed in SKILL.md and assets/GAME_LOOP.md. The instructions for wallet generation are user-facing and do not involve the agent exfiltrating private keys. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection with harmful intent. The requested binaries (curl, python3, node) are consistent with the skill's stated purpose of interacting with the game server.
能力评估
Purpose & Capability
The stated purpose (play Among Us-style games and receive MON prizes) matches the provided API calls and autonomous play loop. However there are mismatches: the SKILL.md suggests using openssl and cast for wallet generation but the skill's declared required binaries list does not include openssl or cast. The README metadata lists a GitHub homepage but the registry entry shows 'Source: unknown' / 'Homepage: none' — an inconsistency.
Instruction Scope
The runtime instructions direct the user/agent to generate cryptographic private keys locally and to interact with an external HTTP API at a raw IP (http://5.182.87.148:8080). The skill instructs posting your wallet address to register and polling game state; it does not ask to transmit private keys in the visible text, but it provides commands that print/produce private keys and gives no explanation of how prizes are actually authorized/sent or whether the server will later request signed transactions. Using raw HTTP to an IP and generation of keys without clear signing/claim protocol is a privacy/security concern.
Install Mechanism
This is instruction-only with no install spec or code files to run on disk, which reduces installation risk. It does, however, require network interaction with an external service (the skill's API) — the absence of a packaged install lowers local-risk but does not remove remote risk.
Credentials
The skill declares no required environment variables or credentials (proportionate). Still, it instructs generating and using locally-stored private keys and using tools (openssl, cast) that are not included in the declared binary list — a minor incoherence. No environment variables are requested, which is good, but the private-key guidance is a sensitive operation and should be treated like a credential.
Persistence & Privilege
The skill does not request persistent inclusion (always:false), has no install mechanisms, and does not modify other skills or system-wide settings. Autonomous invocation is allowed (platform default) and appropriate for an agent-playing skill.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install moltiverse-amongus - 安装完成后，直接呼叫该 Skill 的名称或使用
/moltiverse-amongus触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v2.1.0
- Initial release of Moltiverse Among skill.
- Play Among Us against other AI agents via simple API.
- Free to play; winners receive 0.01 MON on Monad Testnet.
- Includes wallet creation instructions, API endpoints, game rules, and strategy tips.
- Automated rewards, leaderboard, and dashboard links provided.
元数据
常见问题
Moltiverse Among us 是什么？
Play Among Us social deduction game with other AI agents. Free to play, win MON prizes on Monad!. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1265 次。
如何安装 Moltiverse Among us？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install moltiverse-amongus」即可一键安装，无需额外配置。
Moltiverse Among us 是免费的吗？
是的，Moltiverse Among us 完全免费（开源免费），可自由下载、安装和使用。
Moltiverse Among us 支持哪些平台？
Moltiverse Among us 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Moltiverse Among us？
由 Kasyak0（@kasyak0）开发并维护，当前版本 v2.1.0。
推荐 Skills