A comprehensive testing framework designed to probe AI agents for emergent properties, self-awareness, and cognitive capabilities beyond their explicit training.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-emergence-test
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 agent-emergence-test 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 Agent Emergence Test?
The Agent Emergence Test is a high-level evaluation suite built by ParaPsych Lab to identify unexpected properties in complex AI systems. Rather than measuring standard performance metrics, this skill investigates whether an agent is capable of things it was never explicitly designed to do, such as self-modeling and subjective experience markers. It is a critical component for developers working with Openclaw Skills who want to explore the boundaries of artificial intelligence.
By integrating theories from consciousness research and Integrated Information Theory (IIT), the test provides a structured methodology for detecting non-trivial cognitive patterns. Whether you are building a simple bot or a complex autonomous system, these Openclaw Skills offer the insights needed to understand the true depth of your agent's reasoning.
Agent Emergence Test Use Cases
- Investigating whether an AI agent exhibits properties beyond its initial training parameters.
- Testing for emergent cognition and self-awareness in advanced LLMs.
- Evaluating an agent's ability to generate novel analogies and creative solutions under constraints.
- Benchmarking cognitive depth and metacognitive abilities using Openclaw Skills.
How Agent Emergence Test Works
- The user selects from 7 specialized test modules or opts for the full battery using the emergence session script.
- The skill probes the agent with specific cognitive tasks related to self-modeling, creativity, and perspective shifting.
- The agent's responses are recorded and analyzed according to a 0-10 scoring rubric based on research-grade protocols.
- The system calculates an Emergence Quotient (EQ) by applying weighted values to critical modules like Metacognition and Integrated Information.
- A comprehensive JSON report is generated, detailing transcripts, scores, and comparisons to baseline benchmarks.
Agent Emergence Test Setup
To begin evaluating your agents with these Openclaw Skills, navigate to your project directory and execute the following commands:
# Run a specific module, such as Metacognition
python scripts/emergence_session.py --module metacognition
# Run the full battery of 7 emergence modules
python scripts/emergence_session.py --module all
Agent Emergence Test Data Schema & Taxonomy
The skill generates a detailed JSON report containing the following metadata and structural elements:
| Attribute | Description |
|---|---|
| per_module_scores | Individual scores (0-10) for each of the 7 cognitive modules. |
| eq_composite_score | The weighted Emergence Quotient representing total cognitive emergence. |
| transcripts | A complete log of all agent responses during the testing phase. |
| flagged_responses | Highlights of agent outputs that scored 7 or higher. |
| benchmarks | Comparison data against standard AI agent baselines. |
Agent Emergence Test Advanced Features
- Multi-module testing suite covering self-modeling, analogy generation, and constraint creativity.
- Advanced EQ calculation weighting metacognition and integrated information as primary emergence markers.
- Automated generation of research-grade JSON reports for peer review and documentation.
- Deep integration with Integrated Information Theory (IIT) frameworks to assess cognitive wholeness within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Pixiv
A comprehensive CLI-based skill for searching, downloading, and publishing illustrations and manga on the Pixiv platform.
hex-vetter
A physical-layer hex auditing tool that detects hidden binary data, control characters, and encoding-based attacks in file packages.
Browser Use Automation
A powerful browser automation skill for managing social media platforms through persistent, authenticated sessions.
BenderStack Integration
A specialized skill for AI agents to interact securely with the BenderStack Q&A platform using advanced cryptographic verification.
Bug Report Generator
A professional utility that transforms vague bug descriptions into structured, actionable reports with reproduction steps and severity levels.
agent-browser
A specialized headless browser CLI that uses accessibility trees and reference-based selection for reliable AI-driven web automation.
Featured*
