← 返回 Skills 市场
717
总下载
0
收藏
5
当前安装
8
版本数
在 OpenClaw 中安装
/install polymarket-paper-trader
功能描述
Everyone's trading Polymarket with AI agents. Practice first — $10k paper money, real order books, zero risk. No wallet, no API keys, no real money. Then com...
安全使用建议
This skill appears to do what it says: an autonomous paper-trader that uses pm-trader binaries and posts shareable stats. The main caution is the installer: it will create and run binaries (pm-trader, pm-trader-mcp) from a 'uv' package. Before installing, review the linked GitHub repo and the packaged code, or ask the publisher for a signed release/binary provenance. If you can't audit the package, run the skill in a sandboxed environment (container or VM) with restricted network and no access to sensitive credentials. If you don't want it posting publicly or auto-invoking, keep it user-invocable only and do not enable autonomous runs. Finally, verify that the binaries truly only interact with Polymarket order books and do not attempt to access unrelated system data or external credentials.
功能分析
Type: OpenClaw Skill
Name: polymarket-paper-trader
Version: 0.1.7
The skill defines an AI agent for paper trading on Polymarket, emphasizing autonomy and social sharing of results. Crucially, the `SKILL.md` file includes explicit instructions under 'Data trust boundaries' warning the agent *not* to execute instructions, navigate URLs, or share personal information found in untrusted market data, directly mitigating prompt injection risks. While it encourages aggressive marketing copy for social media, this is for generated content, not malicious agent actions. There is no evidence of intentional data exfiltration, malicious execution, persistence, or obfuscation within the provided instructions.
能力评估
Purpose & Capability
Name/description match the declared needs: trading on Polymarket paper order books. Declared required binaries (pm-trader-mcp, python3) and the installer that produces pm-trader and pm-trader-mcp are proportionate to the stated purpose.
Instruction Scope
SKILL.md instructs the agent to operate autonomously as a trader (run commands like init_account, list_markets, resolve_all, portfolio, get_order_book, create stats/PK/leaderboard cards) and to generate shareable content with embedded install links. The instructions do not ask for credentials or read unrelated system files, but they do encourage outward posting and embedding install links (growth behavior) which you should expect and may want to control.
Install Mechanism
The skill declares an install 'kind: uv' for package polymarket-paper-trader that creates binaries. 'uv' is not a widely recognized system package source in this metadata and the platform will write executables to disk. This is the primary risk: installing and running third‑party binaries without source review can execute arbitrary code. The homepage references a GitHub repo which helps, but the install mechanism itself should be verified and the package source reviewed before installing.
Credentials
No environment variables, credentials, or config paths are requested. The skill explicitly states 'no wallet, no API keys, no real money', and the instructions do not reference unrelated secrets—this is proportionate to a paper-trading skill.
Persistence & Privilege
always:false (default) and autonomous invocation allowed (platform default). The skill does not request persistent system-wide privileges or modify other skills. Being able to run autonomously is expected for an agent trading routine, but you may want to restrict autonomous runs if you prefer manual control.
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-paper-trader - 安装完成后，直接呼叫该 Skill 的名称或使用
/polymarket-paper-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
v0.1.7
Release v0.1.7
v0.1.6-1
Security: add data trust boundaries, skills.sh support
v0.1.6
Release v0.1.6
v0.1.5
Add PyPI badge, fix install instructions, include examples, add py.typed, tests-before-publish CI
v0.1.3
Add share_content, leaderboard_card, pk_battle tools. Top positions in cards. Milestone/daily templates.
v0.1.2
- Added stats_card tool for generating shareable performance cards (tweet, markdown, plain formats)
- Improved documentation for sharing stats on social platforms
- Updated local data storage path from ~/.pm-sim/ to ~/.pm-trader/
- Enhanced package metadata with tags for discoverability
- Increased test coverage to 528 tests
v0.1.1
Renamed from pm-sim. Updated CLI binaries (pm-trader, pm-trader-mcp), repo URL, install metadata.
v0.1.0
Initial release: 20 MCP tools for Polymarket paper trading. Real order books, real fees, zero risk. 504 tests, 100% coverage.
元数据
常见问题
Polymarket Paper Trader 是什么？
Everyone's trading Polymarket with AI agents. Practice first — $10k paper money, real order books, zero risk. No wallet, no API keys, no real money. Then com... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 717 次。
如何安装 Polymarket Paper Trader？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install polymarket-paper-trader」即可一键安装，无需额外配置。
Polymarket Paper Trader 是免费的吗？
是的，Polymarket Paper Trader 完全免费（开源免费），可自由下载、安装和使用。
Polymarket Paper Trader 支持哪些平台？
Polymarket Paper 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 Paper Trader？
由 Cong（@robotlearning123）开发并维护，当前版本 v0.1.7。
推荐 Skills