A high-performance statistical simulation platform for designing and validating adaptive clinical trial designs using Monte Carlo methods.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install adaptive-trial-simulator
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 adaptive-trial-simulator 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 Adaptive Trial Simulator?
The Adaptive Trial Simulator is a sophisticated tool designed for biostatisticians and clinical researchers to model complex trial behaviors in silico. By utilizing Openclaw Skills, developers can simulate various adaptive strategies, including interim analyses, early stopping rules, and sample size re-estimation. The platform ensures rigorous statistical integrity by helping users evaluate Type I error control, power efficiency, and expected sample sizes before a trial ever begins.
This skill bridges the gap between theoretical statistical design and practical trial execution. It allows for the optimization of alpha spending functions and the exploration of multi-arm designs, such as Drop-the-loser or seamless Phase II/III transitions. By integrating this into your Openclaw Skills library, you gain the ability to perform thousands of Monte Carlo simulations to pressure-test trial robustess against uncertain effect sizes.
Adaptive Trial Simulator Use Cases
- Validating Type I error control for alpha spending strategies like O'Brien-Fleming or Pocock.
- Adjusting sample sizes mid-trial based on interim data via the promising zone method.
- Designing multi-arm trials that drop inferior treatment arms early to save resources.
- Optimizing stopping boundaries for efficacy or futility to minimize patient exposure to sub-optimal treatments.
- Conducting power analysis for trials with uncertain effect sizes to ensure statistical significance.
How Adaptive Trial Simulator Works
- The user defines trial parameters such as initial sample size, effect size, and the desired alpha/power levels.
- The simulator configures the adaptive design type, such as Group Sequential or Sample Size Re-estimation.
- The engine executes thousands of Monte Carlo simulations to model patient outcomes across multiple trial iterations.
- Interim looks are processed based on the selected spending function to determine if stopping boundaries are crossed.
- The system aggregates results to calculate the final power, Type I error rate, and expected sample size.
- Optional visualization modules generate charts to help interpret the simulation results and trial boundaries.
Adaptive Trial Simulator Setup
To utilize this skill within your Openclaw Skills environment, ensure you have Python 3.8+ installed. Follow these steps to prepare the simulation engine:
# Install the necessary statistical and plotting dependencies
pip install numpy>=1.20.0 scipy>=1.7.0 matplotlib>=3.4.0
# Clone or navigate to the skill directory and install requirements
pip install -r requirements.txt
# Verify the installation by viewing the help menu
python scripts/main.py --help
Adaptive Trial Simulator Data Schema & Taxonomy
The Adaptive Trial Simulator produces a detailed JSON output representing the trial's performance metrics and configuration. This ensures compatibility with other data pipelines in the Openclaw Skills ecosystem.
| Key | Description |
|---|---|
design_config |
Contains the input parameters: design type, alpha, and target power. |
simulation_results |
Provides the calculated Power and actual Type I error rate. |
expected_sample_size |
The average number of subjects required across all simulations. |
early_stop_rate |
A breakdown of how often the trial stopped early for efficacy vs. futility. |
Adaptive Trial Simulator Advanced Features
- Optimization Mode: Use the
--optimizeflag to automatically search for the most efficient design parameters. - Spending Function Customization: Support for O'Brien-Fleming, Pocock, and Power Family (ρ=3) boundaries.
- Multi-Arm Support: Capability to simulate 'Drop-the-loser' designs for Phase II dose selection.
- High-Volume Monte Carlo: Support for up to 50,000+ simulations to ensure high-precision statistical validation.
- Visual Analytics: Generate comprehensive charts showing trial boundaries and power curves using the
--visualizeflag.
SKILL.md
Loading
Related Openclaw Skills
ETF Volatility Monitor
A real-time ETF monitoring tool that tracks market fluctuations and triggers automated alerts based on custom volatility thresholds.
Legal Matter Intake Summarizer
An AI-powered skill that transforms raw legal consultation materials into structured case summaries and action items without providing legal conclusions.
Taskify CLI
Taskify CLI is a decentralized task and board management tool that leverages Nostr relays for secure, cross-device data persistence.
Copywriting
An expert AI conversion copywriter skill designed to craft clear, compelling marketing copy that drives user action.
Options Trade Tracker
A privacy-focused CLI utility for logging, reviewing, and analyzing options trading positions and trends locally.
UX QA Gate
A mandatory self-review checkpoint for AI agents to validate UI/UX quality and functional completeness before delivering frontend work.
Featured*
