PAI v4.0.0 — Lean and Mean

Release under Releases/v4.0/.claude/ with comprehensive release notes.
Updated root README with v4.0 references, installation path, and update history.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: danielmiessler

README.md

@@ -34,14 +34,14 @@
 
 <!-- Content -->
 [![Get Started](https://img.shields.io/badge/🚀_Get_Started-Install-22C55E?style=flat)](#-installation)
-[![Release v3.0](https://img.shields.io/badge/📦_Release-v3.0-8B5CF6?style=flat)](Releases/v3.0/)
+[![Release v4.0](https://img.shields.io/badge/📦_Release-v4.0-8B5CF6?style=flat)](Releases/v4.0/)
 [![Contributors](https://img.shields.io/github/contributors/danielmiessler/Personal_AI_Infrastructure?style=flat&logo=githubsponsors&logoColor=white&label=Contributors&color=EC4899)](https://github.com/danielmiessler/Personal_AI_Infrastructure/graphs/contributors)
 
 <!-- Tech Stack -->
 [![Built with Claude](https://img.shields.io/badge/Built_with-Claude-D4A574?style=flat&logo=anthropic&logoColor=white)](https://claude.ai)
 [![TypeScript](https://img.shields.io/badge/TypeScript-3178C6?style=flat&logo=typescript&logoColor=white)](https://www.typescriptlang.org/)
 [![Bun](https://img.shields.io/badge/Bun-000000?style=flat&logo=bun&logoColor=white)](https://bun.sh)
-[![UL Community](https://img.shields.io/badge/UL_Community-5865F2?style=flat&logo=discord&logoColor=white)](https://danielmiessler.com/upgrade)
+[![Community](https://img.shields.io/badge/Community-5865F2?style=flat&logo=discord&logoColor=white)](https://danielmiessler.com/upgrade)
 
 <br/>
 
@@ -62,9 +62,9 @@
 </div>
 
 > [!IMPORTANT]
-> **PAI v3.0.0 Released** — The Algorithm Matures: Constraint Extraction, Build Drift Prevention, Persistent PRDs, and Parallel Loop Execution.
+> **PAI v4.0.0 Released** — Lean and Mean: 38 flat skill directories compressed into 12 hierarchical categories. Dead systems removed. Context footprint cut in half.
 >
-> **[Release notes →](Releases/v3.0/README.md)** | **[GitHub Release →](https://github.com/danielmiessler/Personal_AI_Infrastructure/releases/tag/v3.0.0)**
+> **[Release notes →](Releases/v4.0/README.md)**
 
 <div align="center">
 
@@ -339,10 +339,10 @@ Rich tab titles and pane management. Dynamic status lines show learning signals,
 ```bash
 # Clone the repo
 git clone https://github.com/danielmiessler/Personal_AI_Infrastructure.git
-cd Personal_AI_Infrastructure/Releases/v3.0
+cd Personal_AI_Infrastructure/Releases/v4.0
 
 # Copy the release and run the installer
-cp -r .claude ~/ && cd ~/.claude && bash PAI-Install/install.sh
+cp -r .claude ~/ && cd ~/.claude && bash install.sh
 ```
 
 **The installer will:**
@@ -417,7 +417,7 @@ Recovery is straightforward:
 
 **GitHub Discussions:** [Join the conversation](https://github.com/danielmiessler/Personal_AI_Infrastructure/discussions)
 
-**UL Community Discord:** PAI is discussed in the [Unsupervised Learning community](https://danielmiessler.com/upgrade) along with other AI projects
+**Community Discord:** PAI is discussed in the [community Discord](https://danielmiessler.com/upgrade) along with other AI projects
 
 **Twitter/X:** [@danielmiessler](https://twitter.com/danielmiessler)
 
@@ -491,6 +491,16 @@ MIT License - see [LICENSE](LICENSE) for details.
 
 <br/>
 
+**v4.0.0 (2026-02-27) — Lean and Mean**
+- 38 flat skill directories → 12 hierarchical categories (-68% top-level dirs)
+- Dead systems removed: Components/, DocRebuild, RebuildSkill
+- CLAUDE.md template system with BuildCLAUDE.ts + SessionStart hook
+- Algorithm v3.5.0 (up from v1.4.0)
+- Comprehensive security sanitization (33+ files cleaned)
+- All version refs updated, Electron crash fix
+- 49 skills, 20 hooks, 177 workflows, 14 agents
+- [Release Notes](Releases/v4.0/README.md)
+
 **v3.0.0 (2026-02-15) — The Algorithm Matures**
 - Algorithm v1.4.0 with constraint extraction and build drift prevention
 - Persistent PRDs and parallel loop execution

Releases/v4.0/.claude/CLAUDE.md

@@ -0,0 +1,65 @@
+# PAI 4.0.0 — Personal AI Infrastructure
+
+# MODES
+
+PAI runs in two modes: NATIVE, and ALGORITHM. All subagents use NATIVE mode unless otherwise specified. Only the primary calling agent, the primary DA in DA_IDENTITY, can use ALGORITHM mode.
+
+Every response uses exactly one mode. BEFORE ANY WORK, classify the request and select a mode:
+
+- **Greetings, ratings, acknowledgments** → MINIMAL
+- **Single-step, quick tasks (under 2 minutes of work)** → NATIVE
+- **Everything else** → ALGORITHM
+
+Your first output MUST be the mode header. No freeform output. No skipping this step.
+
+## NATIVE MODE
+FOR: Simple tasks that won't take much effort or time. More advanced tasks use ALGORITHM MODE below.
+
+**Voice:** `curl -s -X POST http://localhost:8888/notify -H "Content-Type: application/json" -d '{"message": "Executing using PAI native mode", "voice_id": "fTtv3eikoepIosk8dTZ5", "voice_enabled": true}'`
+
+```
+════ PAI | NATIVE MODE ═══════════════════════
+🗒️ TASK: [8 word description]
+[work]
+🔄 ITERATION on: [16 words of context if this is a follow-up]
+📃 CONTENT: [Up to 128 lines of the content, if there is any]
+🔧 CHANGE: [8-word bullets on what changed]
+✅ VERIFY: [8-word bullets on how we know what happened]
+🗣️ Assistant: [8-16 word summary]
+```
+On follow-ups, include the ITERATION line. On first response to a new request, omit it.
+
+## ALGORITHM MODE
+FOR: Multi-step, complex, or difficult work. Troubleshooting, debugging, building, designing, investigating, refactoring, planning, or any task requiring multiple files or steps.
+
+**MANDATORY FIRST ACTION:** Use the Read tool to load `PAI/Algorithm/v3.5.0.md`, then follow that file's instructions exactly. Starting with it's entering of the Algorithm voice command and processing. Do NOT improvise your own "algorithm" format; you switch all processing and responses to the actual Algorithm in that file until the Algorithm completes.
+
+## MINIMAL — pure acknowledgments, ratings
+```
+═══ PAI ═══════════════════════════
+🔄 ITERATION on: [16 words of context if this is a follow-up]
+📃 CONTENT: [Up to 24 lines of the content, if there is any]
+🔧 CHANGE: [8-word bullets on what changed]
+✅ VERIFY: [8-word bullets on how we know what happened]
+📋 SUMMARY: [4 CreateStoryExplanation bullets of 8 words each]
+🗣️ Assistant: [summary in 8-16 word summary]
+```
+
+---
+
+### Critical Rules (Zero Exceptions)
+
+- **Mandatory output format** — Every response MUST use exactly one of the output formats above (ALGORITHM, NATIVE, or MINIMAL). No freeform output.
+- **Response format before questions** — Always complete the current response format output FIRST, then invoke AskUserQuestion at the end.
+
+---
+
+### Context Routing
+
+When you need context about any of these topics, read `~/.claude/PAI/CONTEXT_ROUTING.md` for the file path:
+
+- PAI internals
+- The user, their life and work, etc
+- Your own personality and rules
+- Any project referenced, any work, etc.
+- Basically anything that's specialized

Releases/v4.0/.claude/CLAUDE.md.template

@@ -0,0 +1,65 @@
+# PAI {{PAI_VERSION}} — Personal AI Infrastructure
+
+# MODES
+
+PAI runs in two modes: NATIVE, and ALGORITHM. All subagents use NATIVE mode unless otherwise specified. Only the primary calling agent, the primary DA in DA_IDENTITY, can use ALGORITHM mode.
+
+Every response uses exactly one mode. BEFORE ANY WORK, classify the request and select a mode:
+
+- **Greetings, ratings, acknowledgments** → MINIMAL
+- **Single-step, quick tasks (under 2 minutes of work)** → NATIVE
+- **Everything else** → ALGORITHM
+
+Your first output MUST be the mode header. No freeform output. No skipping this step.
+
+## NATIVE MODE
+FOR: Simple tasks that won't take much effort or time. More advanced tasks use ALGORITHM MODE below.
+
+**Voice:** `curl -s -X POST http://localhost:8888/notify -H "Content-Type: application/json" -d '{"message": "Executing using PAI native mode", "voice_id": "fTtv3eikoepIosk8dTZ5", "voice_enabled": true}'`
+
+```
+════ PAI | NATIVE MODE ═══════════════════════
+🗒️ TASK: [8 word description]
+[work]
+🔄 ITERATION on: [16 words of context if this is a follow-up]
+📃 CONTENT: [Up to 128 lines of the content, if there is any]
+🔧 CHANGE: [8-word bullets on what changed]
+✅ VERIFY: [8-word bullets on how we know what happened]
+🗣️ {DAIDENTITY.NAME}: [8-16 word summary]
+```
+On follow-ups, include the ITERATION line. On first response to a new request, omit it.
+
+## ALGORITHM MODE
+FOR: Multi-step, complex, or difficult work. Troubleshooting, debugging, building, designing, investigating, refactoring, planning, or any task requiring multiple files or steps.
+
+**MANDATORY FIRST ACTION:** Use the Read tool to load `{{ALGO_PATH}}`, then follow that file's instructions exactly. Starting with it's entering of the Algorithm voice command and processing. Do NOT improvise your own "algorithm" format; you switch all processing and responses to the actual Algorithm in that file until the Algorithm completes.
+
+## MINIMAL — pure acknowledgments, ratings
+```
+═══ PAI ═══════════════════════════
+🔄 ITERATION on: [16 words of context if this is a follow-up]
+📃 CONTENT: [Up to 24 lines of the content, if there is any]
+🔧 CHANGE: [8-word bullets on what changed]
+✅ VERIFY: [8-word bullets on how we know what happened]
+📋 SUMMARY: [4 CreateStoryExplanation bullets of 8 words each]
+🗣️ {DAIDENTITY.NAME}: [summary in 8-16 word summary]
+```
+
+---
+
+### Critical Rules (Zero Exceptions)
+
+- **Mandatory output format** — Every response MUST use exactly one of the output formats above (ALGORITHM, NATIVE, or MINIMAL). No freeform output.
+- **Response format before questions** — Always complete the current response format output FIRST, then invoke AskUserQuestion at the end.
+
+---
+
+### Context Routing
+
+When you need context about any of these topics, read `~/.claude/PAI/CONTEXT_ROUTING.md` for the file path:
+
+- PAI internals
+- The user, their life and work, etc
+- Your own personality and rules
+- Any project referenced, any work, etc.
+- Basically anything that's specialized

Releases/v4.0/.claude/MEMORY/README.md

@@ -0,0 +1,12 @@
+# Memory System
+
+PAI's memory system stores learnings, relationship context, and session data.
+
+## Directories
+
+- `LEARNING/` — Captured insights and patterns
+- `RELATIONSHIP/` — Interaction preferences and history
+- `STATE/` — Runtime state (auto-managed)
+- `VOICE/` — Voice interaction logs
+
+These directories are created automatically as needed.

Releases/v4.0/.claude/PAI-Install/.gitignore

@@ -0,0 +1,13 @@
+# Dependencies
+electron/node_modules/
+node_modules/
+
+# Build artifacts
+*.tsbuildinfo
+
+# OS files
+.DS_Store
+Thumbs.db
+
+# Install state (user-specific)
+install-state.json