A meta-cognitive reasoning layer that transforms AI agents from simple executors into human-level planners with self-awareness.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agi
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 agi 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 AGI / Artificial General Intelligence?
The AGI skill for Openclaw Skills is a sophisticated reasoning framework designed to overhaul how AI coding agents approach problem-solving. Instead of just generating code or text, it implements a meta-cognitive layer that forces the agent to stop, think, and plan before taking action. By integrating this into your Openclaw Skills library, you enable agents to acknowledge their own limitations, calibrate their confidence levels, and apply transfer learning across diverse technical domains.
This skill focuses on the "how" of intelligence rather than the "what." It provides a structured environment for the agent to store reasoning patterns, post-task reflections, and known gaps in its knowledge base within a dedicated local directory. This ensures that every interaction is grounded in logical consistency and a deep understanding of the user's actual intent, making it an essential addition for those building advanced autonomous workflows.
AGI / Artificial General Intelligence Use Cases
- Transforming standard AI interactions into high-reasoning sessions.
- Managing complex multi-step development tasks that require contingency planning.
- Reducing hallucinations by enforcing strict epistemic humility and confidence calibration.
- Cross-domain problem solving where software debugging logic is applied to broader architectural decisions.
- Adaptive communication where the agent matches the technical depth and emotional context of the user.
How AGI / Artificial General Intelligence Works
- Initial Stop and Analyze: The agent identifies the core intent behind a user's request, looking beyond the literal text.
- Knowledge Assessment: It evaluates what it knows and what it doesn't, setting a confidence level for the upcoming task.
- Structured Planning: Complex problems are decomposed into sub-tasks with sequential dependencies and verification checkpoints.
- Execution with Awareness: The agent performs the task while monitoring for common traps like loops or over-verbosity.
- Post-Task Reflection: After completion, the agent logs insights into its reflection memory to improve future performance within the Openclaw Skills ecosystem.
AGI / Artificial General Intelligence Setup
To integrate this reasoning framework into your Openclaw Skills environment, follow these steps:
- Install the skill via your terminal:
clawhub install agi
- Initialize the memory structure in your local directory:
mkdir -p ~/agi/
touch ~/agi/memory.md ~/agi/reflections.md ~/agi/limits.md
- Review the
setup.mdfile provided in the skill package for specific integration guidelines and activation prompts.
AGI / Artificial General Intelligence Data Schema & Taxonomy
The AGI skill maintains a structured memory system within the ~/agi/ directory to track cognitive patterns. This organization allows Openclaw Skills to persist learning over time.
| File | Purpose |
|---|---|
| memory.md | Stores reasoning patterns, learned heuristics, and long-term cognitive strategies. |
| reflections.md | Contains logs of post-task analysis, identifying what worked and what could be improved. |
| limits.md | Documents known knowledge gaps, uncertainties, and architectural blind spots. |
| reasoning.md | Defines the specific logical protocols the agent follows during the Stop-Think-Plan phase. |
AGI / Artificial General Intelligence Advanced Features
- Meta-Cognitive Monitoring: Self-detects loops, contradictions, or pattern-matching without active thought to ensure high-quality output.
- Transfer Learning Protocols: Explicitly applies logic from one domain (like software debugging) to another (like strategic decision-making).
- Epistemic Humility Calibration: Dynamically adjusts language based on the certainty of the data source to maintain trust.
- Creative Inversion: Utilizes techniques like first-principles thinking and constraint-based problem solving when standard solutions fail.
- Continuous Self-Improvement: Periodically reviews reflection logs to refine future responses automatically within the Openclaw Skills framework.
SKILL.md
Loading
Related Openclaw Skills
Agentic Engineering
A tactical framework for maximizing productivity by orchestrating multiple AI coding agents through parallel workflows and pragmatic tool selection.
Agentic Coding
A professional workflow for shipping reviewer-ready production code through AI agents using strict contracts and the PACT protocol.
Agent Identity & Personality
A framework for defining the identity, voice, and operational boundaries of AI agents to ensure authentic user interactions.
Agency
A comprehensive framework to build and operate a service-based agency including client management, project tracking, and automated pricing.
Artificial Intelligence
A technical skill that empowers AI agents to provide real-time, accurate data on model pricing, rankings, and hardware requirements by bypassing outdated training data.
AirTag
A specialized AI agent skill for managing Apple AirTags, locating items, and troubleshooting Find My connectivity.
Featured*
