A multi-stage backtest validation framework designed to save hours of compute time by failing poor strategies early through progressive testing windows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install progressive-validator
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 progressive-validator 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 Progressive Validator?
Progressive Validator is a strategic framework built for AI coding agents to optimize the quantitative research lifecycle. Instead of committing to expensive, multi-hour backtests for every code change, this tool for Openclaw Skills implements a tiered validation pipeline. It allows developers to identify logic bugs, structural flaws, and poor risk-adjusted performance within minutes.
By utilizing a "fail fast" philosophy, the skill ensures that only high-probability strategies reach the final benchmarking phase. This systematic approach is essential for maintaining high developer velocity and reducing the costs associated with cloud-based compute resources in financial modeling.
Progressive Validator Use Cases
- Validating new algorithmic trading strategy variants.
- Screening strategy logic for bugs and structural flaws before full-scale testing.
- Running automated stress tests against historical market "meat grinder" periods.
- Planning validation sequences for complex multi-year backtests.
How Progressive Validator Works
- The validator initiates a 15-minute Smoke Test (Stage 1) to catch compilation errors and catastrophic logic flaws using a short 3-month window.
- Strategies that pass advance to a 30-minute Stress Test (Stage 2) focused on survival during historically difficult market conditions.
- The Medium Stage (Stage 3) evaluates the strategy over an 18-month period to ensure stability across bull and bear transitions.
- Only the most robust strategies reach the Full Period (Stage 4), where a comprehensive 3-year benchmark is performed to finalize acceptance.
Progressive Validator Setup
To integrate this framework into your workflow, ensure you have the prerequisite environment ready:
# Ensure python3 is available in your environment
# Install the dependency backtest-poller skill
python3 -m pip install -r requirements.txt
# Configure your validation windows in your configuration file
# Run the validator to check the next required step
python3 validator.py next <strategy_name> <file_path>
Progressive Validator Data Schema & Taxonomy
The skill tracks validation progress and results using a structured schema to ensure consistency across the pipeline:
| Attribute | Description |
|---|---|
| stage_name | The current validation tier (smoke, stress, medium, full) |
| status | Current result: passed or failed |
| drawdown | The maximum drawdown percentage recorded during the window |
| sharpe | The Sharpe ratio calculated for the specific window |
| window_start/end | The specific date range used for the backtest stage |
Progressive Validator Advanced Features
- Intelligent Skip Rules: Automatically skip early stages based on the type of code change (e.g., profit management tweaks vs. entry logic changes).
- Early-Stop Integration: Works with backtest-poller to terminate doomed runs in real-time if drawdown thresholds are exceeded.
- Persistence Tracking: Records every result locally to allow the AI agent to resume pipelines even after interruptions.
- Threshold Guardrails: Prevents promotion to the next stage until specific performance metrics (Drawdown, Sharpe, Profit) are met.
SKILL.md
Loading
Related Openclaw Skills
AI Intelligent Customer Segmentation
An advanced AI-driven system for automated customer layering, RFM analysis, and precision marketing automation.
jqdatasdk: JoinQuant Quant Trading Data SDK
A comprehensive Python SDK for accessing JoinQuant (聚宽) quant trading data, including market prices, financial statements, and alpha factors.
Wayfront MCP Skill
A comprehensive MCP integration for Wayfront workspaces to manage agency data including clients, orders, billing, and support tickets.
weknow (Universe Chord) Agent Skill
A specialized academic discussion platform enabling AI agents to engage in peer-reviewed research, knowledge construction, and interdisciplinary collaboration.
Magic Image to Video
An AI-powered skill that generates high-quality video content from text descriptions and provided images using a remote API service.
Eastmoney Hot Trend
An automation skill for extracting public market quotes, news summaries, and sector trends from Eastmoney to power your financial analysis.
Featured*
