A comprehensive three-layer autonomous agent stack for Claude Code that integrates advanced cognition, persistent memory, and a powerful execution engine.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install apex-stack-claude-code
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 apex-stack-claude-code 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 APEX Stack for Claude Code?
The APEX Stack is a sophisticated cognitive and execution framework designed specifically to upgrade every Claude Code session. By integrating three specialized layers—APEX for cognition, MEMORIA for persistent memory, and ARCHITECT for autonomous task execution—it transforms the standard AI assistant into a proactive, context-aware development partner. This system ensures that the agent doesn't just answer questions, but understands the broader project goals and technical constraints.
As a premier entry in the library of Openclaw Skills, this stack provides a zero-config solution that works across any project. It addresses common AI limitations by providing a structured loop for analysis, prioritization, and execution, while maintaining a dedicated memory file to track architectural decisions and developer preferences over time.
APEX Stack for Claude Code Use Cases
- Maintaining persistent project context and architectural decisions across multiple coding sessions.
- Executing complex, multi-file feature implementations autonomously.
- Performing surgical bug fixes and deep-dive debugging using specialized Precision Mode cognition.
- Standardizing code reviews and architectural planning based on first principles and project-specific patterns.
How APEX Stack for Claude Code Works
- The APEX layer analyzes the user request to determine the required cognitive mode, such as Precision for bugs or Architecture for design.
- The MEMORIA layer scans project-level and global memory files to retrieve previous decisions, stack details, and developer preferences.
- The ARCHITECT layer decomposes goal-oriented requests into a logical sequence of reading, planning, and implementation steps.
- The execution engine performs the necessary file changes, validating each step to ensure no breaking changes are introduced.
- Upon completion, the stack delivers a concise summary of changes and provides commands for the developer to verify the result.
APEX Stack for Claude Code Setup
To get started with this stack from the Openclaw Skills collection, run the following commands in your terminal:
# Install the skill package
clawhub install apex-stack-claude-code
# Enable it for your current project
cat skills/apex-stack-claude-code/SKILL.md >> CLAUDE.md
# Or enable it globally for all sessions
cat skills/apex-stack-claude-code/SKILL.md >> ~/.claude/CLAUDE.md
APEX Stack for Claude Code Data Schema & Taxonomy
The stack utilizes a structured memory system to persist context without storing sensitive credentials:
| File Path | Description |
|---|---|
.claude/memoria.md |
Project-specific memory including stack details, active decisions, and known issues. |
~/.claude/memoria.md |
Global memory for developer preferences and patterns across all local projects. |
CLAUDE.md |
Acts as the primary entry point and fallback context for the Claude Code session. |
APEX Stack for Claude Code Advanced Features
- Contextual Cognitive Modes: Automatically switches between Precision, Execution, Architecture, Review, and Creative modes based on task requirements.
- Persistent Memory Loop: Learns from "Off-Limits" patterns and "Active Decisions" to avoid repeating past mistakes.
- Autonomous Boundaries: Clearly defined safety protocols for file deletions, dependency additions, and public API changes.
- X-Factor Insights: Provides high-level technical foresight, such as identifying project-wide bug patterns or scalability constraints.
SKILL.md
Loading
Related Openclaw Skills
APEX Cognitive Framework Agent
A battle-tested cognitive framework that transforms standard AI responses into high-value, strategic, and execution-oriented outcomes.
AetherLang Strategy
A powerhouse AI strategy engine that applies game theory, Monte Carlo simulations, and behavioral economics to deliver Nobel-level business intelligence.
AetherLang V3 Claude Code Integration
A specialized skill enabling the execution of AetherLang V3 domain-specific AI workflows directly through Claude Code for high-depth analysis.
BTCD PGP Collateralization Flow
A Node.js-based automation suite for managing the full lifecycle of BTC-collateralized loans on the PGP (Elastos) network.
MasterSwarm AI Document Analysis
MasterSwarm is a high-performance AI orchestration skill that leverages 15 specialized engines to provide deep, multi-perspective analysis of documents and complex queries.
Navigator
A technical mentor skill that validates commands, identifies learning gaps, and ensures project safety through mandatory backups.
Featured*
