← Back to Skills Marketplace
patches429

Storyclaw Polymarket Trading

by Parker · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
290
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install storyclaw-polymarket-trading
Description
Self-evolving Polymarket trading bot. Design strategy with user, run paper trading, auto-improve until edge target met, then ask to switch to live.
Usage Guidance
Key things to check before installing or using this skill: - Credentials: Inspect the setup and load_config logic. The metadata declares POLYMARKET_PRIVATE_KEY but most scripts expect credentials/{USER_ID}.json. Don't blindly paste your private key into the environment or a file until you confirm which path the code will actually read. Prefer creating a per-user credentials file with restricted permissions (the code sets 0600 when saving). - Keep it dry: The skill defaults to dry-run, and SKILL.md states never to go live without explicit confirmation. Use dry runs and review recorded performance (state/*.perf.json) before any live activation. - Crontab changes: SKILL.md contains shell snippets that append cron jobs to your crontab. Those commands will alter your system crontab if you run them. Only install cron entries you understand and verify the exact command and paths. Consider running scheduled jobs under a restricted user or via a container instead. - Undeclared dependency: The scripts call 'openclaw notify' for notifications but 'openclaw' is not listed as a required binary. Verify what 'openclaw' does in your environment or remove those calls if you don't want external notifications. - Network endpoints: The code contacts gamma-api.polymarket.com and clob.polymarket.com (Polymarket-related endpoints). Confirm those are the expected endpoints. If you need to audit network traffic, run the code in a controlled environment (container, VM) and monitor outbound requests. - Review setup flow: The truncated setup code may derive and store API keys/secrets locally. Inspect the full setup routine to ensure it does not transmit your private key or derived secrets to any unexpected remote server. - Least privilege & backups: Only give the minimum credentials required for testing (consider a funded test account / small balance). Keep backups of your private key and revoke/rotate keys if you suspect misuse. - If unsure, run the scripts locally in an isolated environment (Docker/VM) and step through 'check' and 'fast' commands before enabling any cron or live trading. If you want, provide the missing setup.py/truncated parts for a deeper review.
Capability Analysis
Type: OpenClaw Skill Name: storyclaw-polymarket-trading Version: 0.1.0 The skill bundle implements a functional Polymarket trading bot with support for both dry-run (simulated) and live trading. It includes comprehensive logic for market scanning (market_scanner.py), signal analysis (signals.py), and automated strategy improvement (strategy_manager.py). While the bot handles sensitive information like private keys and modifies system crontabs for scheduling, these actions are transparently documented in SKILL.md and are necessary for the stated purpose; the code also includes security best practices such as restricted file permissions (0o600) for credential storage in polymarket.py.
Capability Assessment
Purpose & Capability
Name/description match the included scripts: the code implements market scanning, signals, dry-run settlement, review/auto-improve and live trade flows against Polymarket/Gamma/CLOB APIs. However the registry/metadata declares POLYMARKET_PRIVATE_KEY as the required env var / primary credential, while most runtime code expects per-user credential files (credentials/{USER_ID}.json) and uses USER_ID/TELEGRAM_USER_ID to find them. That mismatch is not fatal but is inconsistent and may confuse users about how to supply keys.
Instruction Scope
SKILL.md and scripts stay within trading/strategy scope, and include explicit safeguards (dry-run default, require confirmation before live). The documentation shows shell commands that will modify the user's crontab (echo into crontab) and run pip installs; those are powerful operations that the skill does not perform autonomously but instructs the operator to run. The runtime code also invokes an external CLI 'openclaw notify' for notifications (fallbacks to print), but 'openclaw' is not declared as a required binary in the metadata.
Install Mechanism
No install spec / no remote downloads are present in the registry metadata (instruction-only). The repository includes Python scripts that will be run locally; nothing in the manifest fetches arbitrary archives or runs installers automatically.
Credentials
The skill requests a private key (POLYMARKET_PRIVATE_KEY) which is proportionate for a trading bot, but the code predominantly reads per-user credentials files (credentials/{USER_ID}.json) and derives/stores API keys/secret/passphrase. The declared primaryEnv (POLYMARKET_PRIVATE_KEY) is not consistently referenced in the code paths shown, and the SKILL.md suggests both approaches — this inconsistency could lead users to supply keys in an unexpected location. The skill will store keys/config locally (credentials/ .json) and uses network endpoints (gamma-api.polymarket.com, clob.polymarket.com) which are appropriate for the described purpose.
Persistence & Privilege
The skill does not set always:true. It does instruct the user how to create crontab entries that will run the scripts regularly (i.e., persistence via cron is user-driven). The scripts persist state, strategies, credentials, and performance under the skill directory, which is expected for this functionality. Autonomous invocation by the agent is allowed by default (disable-model-invocation=false) — normal for skills but increases blast radius combined with credential access; no evidence the skill modifies other skills or global agent settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install storyclaw-polymarket-trading
  3. After installation, invoke the skill by name or use /storyclaw-polymarket-trading
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Initial release: Self-evolving Polymarket trading bot with user-driven strategy design and automated paper trading. - Guides user through secure setup and strategy creation for any Polymarket type. - Runs paper trading by default, auto-improves strategy parameters based on performance. - Only switches to live trading after explicit user confirmation and edge target is met. - Multi-user support via per-user credential files and environment variables. - Provides clear rules to prevent unauthorized actions and limit risk. - Includes modular scripts for strategy management, signal scanning, and automatic review.
Metadata
Slug storyclaw-polymarket-trading
Version 0.1.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Storyclaw Polymarket Trading?

Self-evolving Polymarket trading bot. Design strategy with user, run paper trading, auto-improve until edge target met, then ask to switch to live. It is an AI Agent Skill for Claude Code / OpenClaw, with 290 downloads so far.

How do I install Storyclaw Polymarket Trading?

Run "/install storyclaw-polymarket-trading" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Storyclaw Polymarket Trading free?

Yes, Storyclaw Polymarket Trading is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Storyclaw Polymarket Trading support?

Storyclaw Polymarket Trading is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Storyclaw Polymarket Trading?

It is built and maintained by Parker (@patches429); the current version is v0.1.0.

💬 Comments