A specialized framework for building computerized adaptive testing systems that use Item Response Theory to dynamically adjust question difficulty.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install adaptivetest
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 adaptivetest 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 Testing with IRT?
This skill provides the technical foundation for implementing Computerized Adaptive Tests (CAT). By leveraging Item Response Theory (IRT) models like 1PL, 2PL, and 3PL, developers can create assessments that accurately estimate a user's ability while reducing the number of required questions by up to 50 percent. This implementation of Openclaw Skills focuses on maximizing Fisher Information to ensure every question contributes significantly to the final ability estimate.
Developers can use these patterns to build systems that adapt in real-time. Whether for placement or certification, the goal is to spend testing time where measurement matters most—near the user's specific ability level.
Adaptive Testing with IRT Use Cases
- K-12 student placement and diagnostic assessments.
- High-stakes professional certification exams.
- Real-time progress monitoring in e-learning platforms.
- Psychometric research and calibration of new assessment items.
How Adaptive Testing with IRT Works
- Initialize the student's ability estimate based on a population mean or prior data.
- Select the optimal item from the bank using strategies like Maximum Fisher Information.
- Administer the question and record the student's response.
- Update the ability estimate using methods like Expected A Posteriori (EAP) or Maximum Likelihood Estimation.
- Evaluate stopping rules such as Standard Error thresholds or maximum item count to conclude the test.
Adaptive Testing with IRT Setup
To begin implementing these logic patterns, ensure your environment has the necessary psychometric libraries installed:
pip install catsim girth mirt
For R-based environments, use:
# Run in R console
install.packages(c("mirt", "catR"))
Consult the internal implementation references to map these libraries to the CAT workflow defined within Openclaw Skills.
Adaptive Testing with IRT Data Schema & Taxonomy
The following parameters are essential for calibrating items and tracking student progress within Openclaw Skills adaptive frameworks:
| Parameter | Name | Description | Range |
|---|---|---|---|
| a | Discrimination | How well the item differentiates ability levels | 0.5 to 2.5 |
| b | Difficulty | The ability level where probability of correct is 0.5 | -3 to +3 |
| c | Guessing | Probability of a correct response by random chance | 0.2 to 0.25 |
| theta | Ability | The latent trait estimate of the test taker | -3 to +3 |
| SE | Standard Error | Precision measurement (inverse of Information) | Target < 0.3 |
Adaptive Testing with IRT Advanced Features
- Content balancing to ensure proportional representation across multiple domains.
- Exposure control using Randomesque or Sympson-Hetter methods to prevent item over-usage.
- Multidimensional IRT support for testing multiple distinct skills simultaneously.
- Bayesian priors and EAP estimation to handle extreme response patterns where MLE fails.
SKILL.md
Loading
Related Openclaw Skills
AdaptiveTest
A production-grade adaptive testing API that uses IRT and AI to deliver precise ability estimates and personalized learning paths.
Text-Preserve Multilingual Image Rendering
An advanced AI agent skill that prevents text corruption in AI-generated images by pre-rendering non-Latin characters through a Canvas-based pipeline.
Tomoviee Text-to-Speech
A sophisticated AI voice synthesis skill that converts text into natural-sounding speech with customizable emotions and speed.
eye2byte
Eye2byte provides your AI agent with real-time visual context by capturing screenshots, audio, and screen recordings directly through the Model Context Protocol (MCP).
PolyClaw
An automated trading and market analysis skill for Polymarket on the Polygon network.
Working with Lockdownd (The Orchard)
A powerful toolkit for interacting with iOS devices over WiFi via the Apple Lockdown Protocol for diagnostics, log streaming, and cryptographic research.
METADATA
Featured*
