A 10-dimension weighted scoring framework designed to enforce disciplined trade evaluation and position sizing for prediction markets.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install trade-validation
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 trade-validation 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 Trade Validation Scoring Framework?
The trade-validation skill provides a rigorous, data-driven approach to evaluating binary and discrete outcomes in prediction markets like Polymarket and Kalshi. By leveraging a structured scoring rubric, it removes emotional bias and ensures that every position is backed by a calculated weighted confidence score. This skill is a core component for developers building advanced trading agents using Openclaw Skills.
At its heart, the framework enforces a strict 80% confidence threshold for execution and includes automatic veto rules for any critical weakness. It prioritizes capital preservation through built-in circuit breakers, position sizing limits, and mandatory counter-argument documentation to ensure a comprehensive analysis of every market opportunity.
Trade Validation Scoring Framework Use Cases
- Evaluating high-stakes prediction market trades on Polymarket or Kalshi.
- Standardizing pre-trade checklists for AI-driven trading agents.
- Calculating disciplined position sizes based on multi-dimensional risk assessment.
- Documenting trade rationale and exit strategies for long-term performance tracking.
How Trade Validation Scoring Framework Works
- The user provides market data and research across ten specific dimensions such as Information Edge and Liquidity.
- Each dimension is assigned a score from 1 to 10 based on the provided rubric.
- The framework applies predefined weights (e.g., 18% for Information Edge) to calculate a final Weighted Score.
- A Veto Check is performed; if any single dimension scores below 4/10, the trade is automatically rejected.
- If the score is above 80% and passes all vetoes, a trade tier is assigned (Min, Standard, or Conviction).
- The final evaluation, including mandatory counter-arguments and exit strategies, is logged to a trade journal.
Trade Validation Scoring Framework Setup
To integrate this framework into your environment, ensure your directory structure is prepared for journaling. Use the following commands to initialize the necessary paths for Openclaw Skills:
mkdir -p projects/polymarket/trade-journal/
# Copy the skill definition to your local agent path
cp skills/trade-validation.md ./agent_configs/
Trade Validation Scoring Framework Data Schema & Taxonomy
The skill organizes trade data using a structured Score Card and a chronological Trade Journal. Data is categorized as follows:
| Data Point | Description |
|---|---|
| Weighted Score | The final percentage calculated from the 10-dimension matrix. |
| Dimension Scores | Individual 1-10 scores for factors like Source Quality and Market Efficiency. |
| Veto Status | A boolean flag indicating if any dimension failed the minimum threshold. |
| Trade Tier | Position sizing recommendation based on the score range (80% to 90%+). |
| Journal Logs | Markdown-formatted files stored in projects/polymarket/trade-journal/ for post-resolution analysis. |
Trade Validation Scoring Framework Advanced Features
- Automatic circuit breakers that halt trading for 24 hours if a daily loss limit ($8) is exceeded.
- Mandatory 1-hour cool-down periods following any loss to prevent revenge trading.
- Time-based restrictions to avoid low-liquidity trading windows (12am-7am).
- Integrated counter-argument requirement to force the consideration of falsification criteria for every trade.
SKILL.md
Loading
Related Openclaw Skills
Reporting
A standardized reporting engine for generating structured system audits, revenue tracking, and progress logs within an AI agent workspace.
Prompt Guard
A specialized security layer designed to detect and neutralize prompt injection attacks in untrusted AI agent inputs.
Negotiation
A professional framework for AI agents to negotiate rates, manage scope creep, and secure favorable contract terms using elite psychological tactics.
n8n Automation & Workflow Delivery
A high-efficiency skill for architecting and deploying production-grade n8n workflows using a massive pre-built template library.
Upwork Hunting: Find, Bid, Win Automation Jobs
A strategic automation skill designed to identify, evaluate, and secure high-paying Upwork contracts using a library of over 2,000 n8n templates.
Bing Search
A zero-configuration search utility that enables AI agents to query the web via Bing without requiring API keys.
Featured*
