Right now the most powerful AI setups are being built inside companies for efficiency and profits.

That’s all good, but I think the purpose of technology is to serve humans—not the other way around. These new AI frameworks should be available to everyone, including people not in technology, so that regular people can use it to help them flourish.

That’s what PAI is. It’s the foundation for building a Personal AI System that understands your larger goals and context, gets better over time, and that works for you because it’s yours. Not some generic chatbot. Not some common assistant. A full platform for magnifying yourself and your impact on the world.

Related reading:

• The Real Internet of Things — The vision behind PAI (full book)
• AI’s Predictable Path: 7 Components — Visual walkthrough of where AI is heading

What is PAI?

PAI (Personal AI Infrastructure) is an open-source template for building your own AI-powered operating system. It’s currently built on Claude Code, but designed to be platform-independent — the architecture, skills, and workflows are structured so future migrations to other AI platforms are straightforward.

Component	Description
Skills	Self-contained AI capabilities with routing, workflows, and documentation
Agents	Specialized AI personalities for different tasks (engineer, researcher, designer)
Hooks	Event-driven automation that captures work and manages state
History	Automatic documentation system (UOCS) that captures everything

Tip

Start clean, small, and simple. Build the scaffolding that makes AI reliable.

What’s New in v0.9.0

Big updates! PAI is now fully platform-agnostic — your AI identity, your system.

Feature	Description
📊 Observability Dashboard	Real-time agent monitoring with live charts
🎭 Genericized Identity	Configure your DA name, it flows everywhere
⚙️ Better Configuration	Clear docs for all environment variables

👉 See full changelog

🚀 Quick Start

Choose your platform:

🍎 macOS

1. Clone PAI

git clone https://github.com/danielmiessler/PAI.git ~/PAI

2. Create Symlink

# Remove existing ~/.claude if present (backup first if needed)
[ -d ~/.claude ] && mv ~/.claude ~/.claude.backup
ln -s ~/PAI/.claude ~/.claude

3. Run the Setup Wizard

~/.claude/tools/setup/bootstrap.sh

4. Add Your API Keys

cp ~/.claude/.env.example ~/.claude/.env
nano ~/.claude/.env

5. Start Claude Code

source ~/.zshrc  # Load PAI environment
claude

🐧 Linux

1. Clone PAI

git clone https://github.com/danielmiessler/PAI.git ~/PAI

2. Create Symlink

# Remove existing ~/.claude if present (backup first if needed)
[ -d ~/.claude ] && mv ~/.claude ~/.claude.backup
ln -s ~/PAI/.claude ~/.claude

3. Run the Setup Wizard

~/.claude/tools/setup/bootstrap.sh

4. Add Your API Keys

cp ~/.claude/.env.example ~/.claude/.env
nano ~/.claude/.env

5. Start Claude Code

source ~/.bashrc  # Load PAI environment
claude

🪟 Windows

1. Clone PAI (PowerShell)

git clone https://github.com/danielmiessler/PAI.git $env:USERPROFILE\PAI

2. Create Symlink (Run PowerShell as Administrator)

# Remove existing .claude if present (backup first if needed)
if (Test-Path "$env:USERPROFILE\.claude") { Rename-Item "$env:USERPROFILE\.claude" "$env:USERPROFILE\.claude.backup" }
New-Item -ItemType SymbolicLink -Path "$env:USERPROFILE\.claude" -Target "$env:USERPROFILE\PAI\.claude"

3. Run the Setup Wizard

& "$env:USERPROFILE\.claude\tools\setup\bootstrap.ps1"

4. Add Your API Keys

Copy-Item "$env:USERPROFILE\.claude\.env.example" "$env:USERPROFILE\.claude\.env"
notepad "$env:USERPROFILE\.claude\.env"

5. Start Claude Code

# Restart PowerShell to load environment, then:
claude

Tip

The setup wizard will configure your name, email, AI assistant name, and environment variables to customize to your environment.

📚 For detailed setup, see docs/QUICKSTART.md

📚 Documentation

All documentation lives in the CORE skill (.claude/skills/CORE/):

Document	Description
CONSTITUTION.md	System philosophy, architecture, operating principles
SkillSystem.md	How to create your own skills — the canonical skill structure guide
SKILL.md	Main PAI skill with identity, preferences, quick reference
HookSystem.md	Event-driven automation
HistorySystem.md	Automatic work documentation (UOCS)

Additional Reference

Document	Description
Prompting.md	Prompt engineering patterns
Aesthetic.md	Visual design system
voice-server/README.md	Text-to-speech feedback

🎨 Examples

Explore example skills in .claude/skills/:

Skill	Description
Observability/	Real-time agent monitoring dashboard with WebSocket streaming
BrightData/	Four-tier progressive web scraping with automatic fallback
Fabric/	Native Fabric patterns — 248 patterns run directly in Claude’s context (no CLI needed)
Research/	Multi-source research workflows
Createskill/	Templates for creating new skills

Each skill demonstrates the skills-as-containers pattern with routing, workflows, and self-contained documentation.

Native Fabric Patterns

The Fabric skill now executes patterns natively within Claude Code — no CLI spawning required:

• Your subscription’s power — Patterns run with your Opus/Sonnet model, not Fabric’s configured model
• Full context — Patterns have access to your entire conversation history
• Faster execution — No process spawning overhead
• 248 patterns included — extract_wisdom, summarize, threat modeling, and more

# Update patterns from upstream
.claude/skills/Fabric/tools/update-patterns.sh

Only use fabric CLI for YouTube transcripts (-y) or pattern updates (-U).

🏗️ The Thirteen Founding Principles

PAI is built on 13 foundational principles that define how to build reliable AI infrastructure.

Complete architecture documentation: .claude/skills/CORE/Architecture.md

---

1. Clear Thinking + Prompting is King

The quality of outcomes depends on the quality of thinking and prompts. Before any code, before any architecture—there must be clear thinking.

---

2. Scaffolding > Model

The system architecture matters more than the underlying AI model. A well-structured system with good scaffolding will outperform a more powerful model with poor structure.

---

3. As Deterministic as Possible

Favor predictable, repeatable outcomes over flexibility. Same input → Same output. Always.

---

4. Code Before Prompts

Write code to solve problems, use prompts to orchestrate code. Prompts should never replicate functionality that code can provide.

---

5. Spec / Test / Evals First

Define expected behavior before writing implementation. If you can’t specify it, you can’t test it. If you can’t test it, you can’t trust it.

---

6. UNIX Philosophy

Do one thing well. Compose tools through standard interfaces. Build small, focused tools—compose them for complex operations.

---

7. ENG / SRE Principles

Apply software engineering and site reliability practices to AI systems. AI infrastructure is infrastructure—treat it with the same rigor.

---

8. CLI as Interface

Every operation should be accessible via command line. If there’s no CLI command for it, you can’t script it or test it reliably.

---

9. Goal → Code → CLI → Prompts → Agents

The proper development pipeline for any new feature. Each layer builds on the previous—skip a layer, get a shaky system.

---

10. Meta / Self Update System

The system should be able to improve itself. A system that can’t update itself will stagnate.

---

11. Custom Skill Management

Skills are the organizational unit for all domain expertise. Skills are how PAI scales—each new domain gets its own skill, maintaining organization as the system grows.

---

12. Custom History System

Automatic capture and preservation of valuable work. Memory makes intelligence compound. Without history, every session starts from zero.

---

13. Custom Agent Personalities / Voices

Specialized agents with distinct personalities for different tasks. Personality isn’t decoration—it’s functional.

🛠️ Technology Stack

Category	Choice	Note
Runtime	Bun	NOT Node.js
Language	TypeScript	NOT Python
Package Manager	Bun	NOT npm/yarn/pnpm
Format	Markdown	NOT HTML for basic content
Testing	Vitest	When needed
Voice	ElevenLabs	TTS integration

🔄 Updating PAI

PAI includes an intelligent sideloading system that helps you update while preserving your customizations.

# In Claude Code, run:
/paiupdate    # or just /pa

What happens:

1. Your DA fetches the latest PAI to a staging area (doesn’t touch your files)
2. Analyzes differences between upstream and your customizations
3. Generates a personalized report showing conflicts vs. safe updates
4. You choose what to adopt — your DA handles the merge

Your custom skills, modified hooks, and personalized settings are never blindly overwritten. The system understands that your env.DA, custom environment variables, and personal tweaks are intentional.

💬 Community

Kai and I work hard to address issues and PRs throughout the week — we try not to get too far behind!

Channel	Link
🐛 Issues	Report bugs or request features
💬 Discussions	Ask questions and share ideas
🎥 Video	Watch the full PAI walkthrough
📝 Blog	The Real Internet of Things

📝 Updates

v0.9.1 (2025-12-04) — Setup Script Fix

PAI_DIR Auto-Configuration

• setup.sh now automatically configures PAI_DIR in settings.json with your actual home directory path
• No more manual editing of __HOME__/.claude placeholder
• Clear error messaging if hooks fail due to misconfigured paths

Improved Documentation

• Added _setupNote in settings.json explaining the fix
• Updated _envDocs with troubleshooting guidance
• QUICKSTART.md troubleshooting section for PAI_DIR issues

Bug Fix

• Fixes #110 — Hook failures caused by unexpanded PAI_DIR placeholder v0.9.0 (2025-12-01) — Platform Agnostic Release

This release focuses on making PAI fully portable and fork-friendly. Your AI, your identity, your system.

Observability Dashboard

• Complete real-time agent monitoring at .claude/Observability/
• WebSocket streaming of all agent activity
• Live pulse charts, event timelines, and swim lanes
• Multiple themes (Tokyo Night, Nord, Catppuccin, etc.)
• Security obfuscation for sensitive data

Genericized Agent Identity

• All agent references now use process.env.DA || 'main'
• No more hardcoded names — your DA name flows through the entire system
• Observability dashboard shows your configured identity

Platform-Agnostic Configuration

• Clear separation: settings.json for identity/paths, .env for API keys
• DA (Digital Assistant name) — your AI’s identity
• PAI_DIR — root directory for all configuration
• TIME_ZONE — configurable timezone for timestamps

Skill System Improvements

• Canonical TitleCase file naming throughout
• Standardized skill-workflow-notification script for dashboard detection
• All paths use ${PAI_DIR}/ for location-agnostic installation v0.8.0 (2025-11-25) — Research & Documentation

Research Skill

• Comprehensive research skill with 10 specialized workflows
• Multi-source research with parallel agent execution
• Fabric pattern integration (242+ AI patterns)

Infrastructure

• Path standardization using ${PAI_DIR}/ throughout
• PAI_CONTRACT.md defining core guarantees
• Self-test validation system for health checks
• Protection system for PAI-specific files v0.7.0 (2025-11-20) — Protection & Clarity

PAI Path Resolution System (#112)

• Centralized pai-paths.ts library — single source of truth
• Smart detection with fallback to ~/.claude
• Updated 7 hooks to use centralized paths

PAI vs Kai Clarity (#113)

• PAI_CONTRACT.md — official contract defining boundaries
• Self-test system (bun ${PAI_DIR}/hooks/self-test.ts)
• Clear README section distinguishing PAI from Kai

Protection System

• .pai-protected.json manifest of protected files
• validate-protected.ts script for pre-commit validation
• Pre-commit hook template for automated checks v0.6.5 (2025-11-18) — BrightData Integration

Four-Tier Progressive Web Scraping

• Tier 1: WebFetch (free, built-in)
• Tier 2: cURL with headers (free, more reliable)
• Tier 3: Playwright (free, JavaScript rendering)
• Tier 4: Bright Data MCP (paid, anti-bot bypass) v0.6.0 (2025-11-15) — Major Architecture Update

Repository Restructure

• Moved all configuration to .claude/ directory
• Skills-as-containers architecture
• Three-tier progressive disclosure

Skills System

• Art skill with visual content generation
• Story-explanation skill for narrative summaries
• Create-skill and create-cli meta-skills

Hook System

• Comprehensive event capture system
• Session summary and tool output capture
• Tab title updates

Voice Integration

• Voice server with ElevenLabs TTS
• Session start notifications v0.5.0 and Earlier

v0.5.0 — Foundation

• CORE skill as central context loader
• Constitution defining system principles
• CLI-First Architecture pattern
• Initial skills: Fabric, FFUF, Alex Hormozi pitch

Pre-v0.5.0 — Early Development

• Initial repository setup
• Basic settings.json structure
• Agent personality definitions
• Foundational hook experiments

📜 License

MIT License — see LICENSE for details.

🙏 Acknowledgments

Built on Claude Code by Anthropic.

PAI is the technical foundation for Human 3.0 — a program I created to help people transform into a version of themselves that can thrive in the post-corporate world that’s coming. Human 3.0 means AI-augmented humans who build and control their own AI systems.

Right now, the most sophisticated AI infrastructure exists inside corporations with massive engineering teams. PAI exists to change that. To give individuals the same scaffolding that companies spend millions building.

Your AI, knowing how you work, learning from your patterns, serving your goals — not some corporation’s engagement metrics. That’s what this enables.

---

Start clean. Start small. Build the AI infrastructure you need.

⬆ Back to Top