← 返回 Skills 市场
djdyll

Manual Trade Placement

作者 Dyll · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ suspicious
512
总下载
0
收藏
1
当前安装
8
版本数
在 OpenClaw 中安装
/install polymarket-manual-trade
功能描述
Place manual trades on Polymarket by telling your agent what to bet on. Supports FAK (instant fill at market) and GTC (limit order on the book). Pass a Simme...
安全使用建议
This skill appears to implement Polymarket trades via Simmer and legitimately needs SIMMER_API_KEY and your wallet key, but there are two red flags: (1) the package/registry metadata omitted required env vars while the included files declare SIMMER_API_KEY and WALLET_PRIVATE_KEY — confirm which is accurate before proceeding; (2) manual_trade.py calls dotenv.load_dotenv('/root/.openclaw/.env'), which will read an agent-global .env file and could expose unrelated secrets. Before installing or providing credentials: review the code yourself (or ask the author) and remove the load_dotenv call, run the skill in a sandboxed environment, only provide a wallet key you can rotate or a test wallet, or prefer using the 'sim' venue for testing. If you don't trust the author or can't inspect/modify the code, do not provide your WALLET_PRIVATE_KEY to this skill.
功能分析
Type: OpenClaw Skill Name: polymarket-manual-trade Version: 1.1.0 The skill provides legitimate functionality for placing manual trades on Polymarket via the Simmer API. The code in `manual_trade.py` handles market resolution, price discovery from the Polymarket CLOB, and order execution using the `simmer-sdk`. It requires sensitive environment variables (`SIMMER_API_KEY` and `WALLET_PRIVATE_KEY`) which are necessary for its stated purpose of on-chain trading. The communication is restricted to expected endpoints (api.simmer.markets and clob.polymarket.com), and there is no evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
The skill's name/description (manual trading on Polymarket via Simmer) matches the code and README: it calls Simmer APIs, fetches CLOB book data, and places orders. However the registry metadata shown at the top says 'Required env vars: none' and 'Primary credential: none' while clawhub.json, README.md, and SKILL.md all declare SIMMER_API_KEY and WALLET_PRIVATE_KEY as required; this mismatch is an incoherence that should be resolved before trusting the skill.
Instruction Scope
SKILL.md and manual_trade.py instruct the agent to import markets, fetch live order book data, and place signed trades via simmer-sdk — all expected. But manual_trade.py unconditionally attempts to load a .env file at /root/.openclaw/.env using dotenv.load_dotenv("/root/.openclaw/.env"). Loading an agent-global .env file is unnecessary for the skill's stated usage and risks reading unrelated secrets stored there. The code otherwise calls only expected external endpoints (api.simmer.markets, clob.polymarket.com).
Install Mechanism
This is instruction-only with a shipped Python script; there is no install spec. clawhub.json lists a pip requirement (simmer-sdk). No external arbitrary downloads or archive extraction are present. The absence of a formal install step is low risk, but the skill will require installing simmer-sdk to operate.
Credentials
Requiring SIMMER_API_KEY and WALLET_PRIVATE_KEY is proportionate to placing signed trades, but these are highly sensitive. The code doesn't directly reference WALLET_PRIVATE_KEY but relies on simmer-sdk (which may read env vars or accept a key) — that is reasonable but sensitive. The bigger proportionality concern is the unconditional load of /root/.openclaw/.env, which may grant the skill access to additional secrets beyond the two declared variables. Also the registry metadata inconsistency (required envs omitted at the top) is a red flag.
Persistence & Privilege
always: false and no autostart/always flags are set. The skill does not request system-wide persistence or modify other skills' configurations. Note: the skill can be invoked autonomously by default (platform normal behavior) — combined with access to a private wallet key this increases blast radius, so only grant credentials to trusted skills.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install polymarket-manual-trade
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /polymarket-manual-trade 触发
  4. 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
v1.1.0
Migrate to AgentSkills spec: clawhub.json with tunables, README, automaton reporting
v1.0.6
Cancel order support (--cancel, --cancel-side), GTC sell order cleanup via cancel_market_orders before stoploss sells, best ask/bid sort fix, portable .env loading
v1.0.5
Cancel order support (--cancel, --cancel-side), best ask/bid sort fix, portable .env loading
v1.0.4
Added Simmer website link (simmer.markets)
v1.0.3
Security fix: removed hardcoded /root/.openclaw/.env path — now searches for .env generically
v1.0.2
Fixed CLOB token discovery after URL import (was falling back to hardcoded 0.55 price); added allow_rebuy=True to prevent SDK blocking repeat buys on same market. Both FAK and GTC fully tested and verified.
v1.0.1
Bug fixes: import_from_url now fetches full market after import to get CLOB tokens for price discovery; added allow_rebuy=True to prevent SDK blocking repeat trades on same market
v1.0.0
Initial release: FAK and GTC order support, auto-import from Polymarket URL, live CLOB price discovery
元数据
Slug polymarket-manual-trade
版本 1.1.0
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 8
常见问题

Manual Trade Placement 是什么?

Place manual trades on Polymarket by telling your agent what to bet on. Supports FAK (instant fill at market) and GTC (limit order on the book). Pass a Simme... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 512 次。

如何安装 Manual Trade Placement?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install polymarket-manual-trade」即可一键安装,无需额外配置。

Manual Trade Placement 是免费的吗?

是的,Manual Trade Placement 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Manual Trade Placement 支持哪些平台?

Manual Trade Placement 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 Manual Trade Placement?

由 Dyll(@djdyll)开发并维护,当前版本 v1.1.0。

💬 留言讨论