← 返回 Skills 市场
Polymarket Coffee Trader
作者
diagnostikon
· GitHub ↗
· v0.0.3
· MIT-0
192
总下载
0
收藏
0
当前安装
5
版本数
在 OpenClaw 中安装
/install polymarket-coffee-trader
功能描述
Trades Polymarket coffee markets using three compounding seasonal edges unique to the global coffee market — Brazil frost window mispricing, harvest cycle aw...
安全使用建议
This skill appears to be a real Polymarket trading bot that reasonably needs an API key and a client library, but the package metadata is inconsistent and there is executable code, so proceed with caution. Before installing or providing any credentials: 1) Confirm which env var(s) the runtime actually requires (SIMMER_API_KEY) and ensure you control the API key's permissions (paper-trade vs live-trade scopes). 2) Verify the simmer-sdk package source and inspect trader.py fully to ensure there are no hidden network endpoints, telemetry, or code paths that place live trades without an explicit --live flag. 3) Keep autostart disabled until you are satisfied with behavior; test in a sandbox with a restricted API key that can only paper-trade. 4) Ask the publisher to correct the registry metadata (declare the required env vars and the pip dependency or provide an explicit install spec) — metadata inconsistencies are a red flag. If you are not comfortable auditing the code or supplying an API key, do not install.
功能分析
Type: OpenClaw Skill
Name: polymarket-coffee-trader
Version: 0.0.3
The skill is a legitimate trading bot designed to execute trades on Polymarket coffee markets based on seasonal meteorological patterns. The code in trader.py is well-structured, transparently implements the described strategy, and follows a 'safe by default' approach by requiring an explicit --live flag for real financial transactions. No indicators of data exfiltration, malicious execution, or prompt injection were found.
能力标签
能力评估
Purpose & Capability
The declared purpose (automated trading on Polymarket) legitimately requires an API credential and a client library; clawhub.json lists a required env var SIMMER_API_KEY and a pip dependency simmer-sdk which match that need. However the skill header/registry metadata states "Required env vars: none" and the overall install section claims "No install spec — instruction-only skill." Those three pieces (registry metadata, presence of clawhub.json, and code) are inconsistent and reduce trust.
Instruction Scope
SKILL.md and the visible portion of trader.py stay on-topic: they describe market discovery, conviction sizing, and a clear paper-vs-live trade model (paper by default; explicit --live required to place real trades). The instructions reference only trade-related environment variables (e.g., SIMMER_ENSO_PHASE) and date-derived signals; there are no obvious instructions to read unrelated system files or exfiltrate arbitrary data.
Install Mechanism
The package is marked as "No install spec" but clawhub.json declares a pip dependency (simmer-sdk). That implies an install step will be needed even though the top-level metadata says none. Using pip to install simmer-sdk is reasonable for a trading skill, but the lack of an explicit install spec is an inconsistency that could hide additional install actions. No direct download-from-URL or archive-extraction signs are present.
Credentials
The only secret-like requirement in clawhub.json is SIMMER_API_KEY, which is proportionate for a trading client that needs to authenticate to a trading service. The code also reads many SIMMER_* tunables (MAX_POSITION, MIN_TRADE, ENDO_PHASE, etc.) which are configuration values rather than unrelated credentials. The mismatch is that the skill registry metadata claims no required env vars while the manifest requires SIMMER_API_KEY; this should be clarified before installing.
Persistence & Privilege
No elevated persistence flags were requested: always is false, autostart is false, and disable-model-invocation is false (normal). clawhub.json declares an automaton entrypoint (trader.py) but autostart=false means it won't run automatically unless explicitly enabled—this is appropriate for a trading 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 polymarket-coffee-trader - 安装完成后，直接呼叫该 Skill 的名称或使用
/polymarket-coffee-trader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
v0.0.3
SDK resilience: try/except runt apply_skill_config
v0.0.2
fix: add _client.live=True so --live actually routes to polymarket-venue for real USDC trades
v1.0.2
Republish to refresh registry metadata
v1.0.1
Fix apply_skill_config AttributeError for new Simmer SDK compatibility
v1.0.0
- Initial release of Coffee Seasonal Trader skill, version 1.0.
- Trades Polymarket coffee markets using three seasonal edges: Brazil frost window mispricing, harvest cycle, and ENSO phase effects.
- Applies adaptive conviction scaling and position sizing based on date-derived multipliers specific to coffee’s unique markets.
- Requires no external APIs by default; all logic using internal, openly documented seasonal calendars.
- Includes safety defaults: paper trading mode enabled unless explicitly set to live; no automation/start without user interaction.
- Detailed documentation provided for strategy, signal logic, sizing examples, and suggestions for advanced remixing.
元数据
常见问题
Polymarket Coffee Trader 是什么？
Trades Polymarket coffee markets using three compounding seasonal edges unique to the global coffee market — Brazil frost window mispricing, harvest cycle aw... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 192 次。
如何安装 Polymarket Coffee Trader？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install polymarket-coffee-trader」即可一键安装，无需额外配置。
Polymarket Coffee Trader 是免费的吗？
是的，Polymarket Coffee Trader 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Polymarket Coffee Trader 支持哪些平台？
Polymarket Coffee Trader 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 Polymarket Coffee Trader？
由 diagnostikon（@diagnostikon）开发并维护，当前版本 v0.0.3。
推荐 Skills