An experimental autonomous agent for analyzing Polymarket public data and simulating paper trading with strict LLM cost controls.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install polymarket-autopilot-experimental
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install polymarket-autopilot-experimental using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Polymarket Autopilot (Experimental)?
This skill provides an automated framework for monitoring Polymarket prediction markets without financial risk. It is designed as a read-only experimental tool that observes public market data, filters noise, and performs paper trading simulations. By integrating with Openclaw Skills, it offers a controlled environment for testing AI decision-making in high-volatility environments while adhering to strict budget constraints and execution frequencies.
The skill prioritizes safety and cost-efficiency, operating entirely without wallets or real currency. It leverages a dual-model approach, utilizing OpenAI for data normalization and Anthropic for high-level reasoning, ensuring that every simulated trade is backed by a prudent analysis of market movements.
Polymarket Autopilot (Experimental) Use Cases
- Tracking high-volume Polymarket prediction trends automatically without manual oversight.
- Testing AI-driven market analysis strategies through paper trading without utilizing real wallets or capital.
- Generating concise Italian-language reports on market movements and simulated portfolio performance.
- Evaluating the cost-to-value ratio of using premium LLMs for financial data synthesis within Openclaw Skills.
How Polymarket Autopilot (Experimental) Works
- The agent fetches public data from Polymarket API without requiring authentication or account registration.
- A pre-filter layer discards markets with low volume or those nearing immediate expiration to ensure data quality.
- Selected markets (max 5) are processed by OpenAI for parsing and Anthropic for qualitative ranking.
- The system simulates a paper trading environment with a €50 simulated capital limit and no leverage.
- It performs internal accounting to track LLM token consumption against a strict weekly budget.
- A final report is generated in Italian, providing a net simulated result and a two-line summary.
Polymarket Autopilot (Experimental) Setup
To deploy this skill within your local environment, follow these steps. Ensure your environment variables for OpenAI and Anthropic are active.
# Add the skill to your local repository
openclaw install polymarket-autopilot-experimental
# Configure the execution constraints
# The skill defaults to 1 run every 3 days
export SKILL_BUDGET_LIMIT=2.00
Polymarket Autopilot (Experimental) Data Schema & Taxonomy
The skill maintains a transparent data structure to track performance and cost metrics across every execution of these Openclaw Skills.
| Data Point | Description |
|---|---|
| Observed Markets | Count of public markets analyzed during the cycle |
| Simulation Stats | Percentage and Euro-denominated paper trading results |
| LLM Cost Breakdown | Detailed token expenditure for OpenAI and Anthropic |
| Net Simulated Value | Paper trading profit/loss minus actual LLM costs |
| Agent Commentary | Qualitative prudential notes on market stability |
Polymarket Autopilot (Experimental) Advanced Features
- Automatic budget kill-switch that halts all activity if LLM costs exceed the weekly limit of 2€.
- Adaptive complexity reduction that simplifies prompts if AI costs outweigh simulation gains for two consecutive cycles.
- Dual-model orchestration to balance processing speed (OpenAI) with deep analytical reasoning (Anthropic).
- Built-in rate limiting to enforce a maximum of one execution every three days for sustainable long-term monitoring.
SKILL.md
Loading
Related Openclaw Skills
Agent Swarm Orchestrator
A high-performance orchestration framework that automates the full coding lifecycle from Obsidian task intake to GitLab merge using AI agents.
Her Voice
A high-performance, on-device text-to-speech engine that gives AI agents a natural-sounding voice with ultra-low latency.
Briefing Room
An automated news station that researches global events and produces a conversational 10-minute audio briefing and document every day.
URL to Markdown Converter
A technical tool for fetching web URLs and converting their HTML content into clean, readable Markdown formats optimized for AI agents.
Living Room Air Quality Monitor
An automated solution for tracking, logging, and reporting indoor air quality metrics from IKEA Dirigera sensors into a local SQLite database.
Living Room Smoke Detector
A specialized emergency monitoring skill that turns IKEA Dirigera air sensors into a continuous audio alert system for smoke and fire detection.
Featured*
