← 返回 Skills 市场
Prediction Market Arbitrage
作者
Hagen Hoferichter
· GitHub ↗
· v1.0.0
730
总下载
2
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install prediction-market-arbitrage
功能描述
Orchestrates monitoring, market odds, and execution proxy tools to detect news-market price gaps and emit arbitrage alerts with optional trade plans.
安全使用建议
This skill appears to do what it says: orchestrate monitoring, market lookups, and optional execution guidance. Before installing, consider these steps: (1) Only provide SIMMER_API_KEY if you explicitly want the agent to be able to produce execution plans or run dry-run/live trades; otherwise keep it unset and use alert-only mode. (2) Store SIMMER_API_KEY in a secrets manager or environment that is not broadly accessible, and rotate it regularly. (3) Audit the upstream skills (topic-monitor, polymarket-odds, simmer-weather) that this meta-skill orchestrates — the meta-skill delegates behavior to them. (4) Run the workflow in dry-run mode first and verify that alerts and decision logic behave as expected. (5) Limit agent autonomy and review logs before enabling any live execution path. If you want a stricter review, provide the upstream skills' code or the Simmer SDK docs so I can check exactly how the API key is used and whether any hidden network endpoints are contacted.
功能分析
Type: OpenClaw Skill
Name: prediction-market-arbitrage
Version: 1.0.0
The skill bundle is classified as benign. It serves as a meta-skill to orchestrate existing OpenClaw skills (`topic-monitor`, `polymarket-odds`, `simmer-weather`) for a prediction market arbitrage workflow. While it involves potentially sensitive actions like trade execution, the `SKILL.md` explicitly includes strong guardrails for the AI agent, such as requiring explicit user confirmation before any live trading action, mandating dry-runs, and rejecting stale data. There is no evidence of intentional harmful behavior, data exfiltration, backdoors, or prompt injection designed to subvert the agent's intended function or bypass security measures. The commands provided are standard for managing and interacting with OpenClaw skills.
能力评估
Purpose & Capability
The skill's name and description match the actions in SKILL.md: it orchestrates topic-monitor, polymarket-odds, and simmer-weather for arbitrage alerting. Requiring python3/node/npx is expected because the upstream skills are Python/Node and installation uses npx. One minor mismatch: SIMMER_API_KEY is declared as required even though the workflow explicitly supports an 'alert-only' execution_mode; making the API key mandatory is not strictly necessary for alert-only operation.
Instruction Scope
The SKILL.md confines runtime actions to invoking the three upstream skills, normalizing values, applying hard decision rules, and emitting alerts or execution plans. It does not instruct the agent to read unrelated system files, other environment variables, or external endpoints beyond the described ClawHub skills and Simmer endpoints. It explicitly warns not to auto-trade unless the user asks.
Install Mechanism
This is instruction-only (no install spec or code blobs). The recommended install uses npx clawhub@latest to install upstream skills — a reasonable, traceable approach. No downloads from arbitrary URLs or archive extractions are present in the skill itself.
Credentials
Only one env var is requested: SIMMER_API_KEY. That is expected for the simmer-weather execution proxy, but declaring it as required (rather than optional) is disproportionate if users will run in alert-only mode. SIMMER_API_KEY is a trading credential; supplying it enables live trading via the cited execution proxy if downstream components or operators invoke execution. Users should treat this key as sensitive, store it in a secrets manager, and avoid providing it unless they intend to permit trade execution.
Persistence & Privilege
The skill does not request always:true, does not modify other skills' configs, and does not demand persistent system-wide privileges. user-invocable is true and autonomous invocation is allowed (disable-model-invocation:false) — this is platform-default and acceptable here, but users should be mindful that an agent with this skill could be asked to perform repeated monitoring and, if credentials are supplied, to prepare execution plans.
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 prediction-market-arbitrage - 安装完成后，直接呼叫该 Skill 的名称或使用
/prediction-market-arbitrage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
v1.0.0
prediction-market-arbitrage 1.0.0
- Initial release: Enables real-time arbitrage detection between breaking news and Polymarket prices for CEO/company event markets.
- Orchestrates topic-monitor, polymarket-odds, and simmer-weather skills to compare news confidence with market probabilities.
- Supports automated alert generation and optional live trade execution via Simmer, with robust guardrails for skill path resolution and user confirmation.
- Requires explicit user confirmation for live trades and provides clear, machine-readable trade recommendations and signal payloads.
元数据
常见问题
Prediction Market Arbitrage 是什么？
Orchestrates monitoring, market odds, and execution proxy tools to detect news-market price gaps and emit arbitrage alerts with optional trade plans. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 730 次。
如何安装 Prediction Market Arbitrage？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install prediction-market-arbitrage」即可一键安装，无需额外配置。
Prediction Market Arbitrage 是免费的吗？
是的，Prediction Market Arbitrage 完全免费（开源免费），可自由下载、安装和使用。
Prediction Market Arbitrage 支持哪些平台？
Prediction Market Arbitrage 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 Prediction Market Arbitrage？
由 Hagen Hoferichter（@h4gen）开发并维护，当前版本 v1.0.0。
推荐 Skills