superlocalmemory 2.6.0 → 2.7.0

package/README.md

@@ -30,9 +30,8 @@
 <p align="center">
   <a href="https://superlocalmemory.com"><strong>superlocalmemory.com</strong></a> •
   <a href="#-quick-start">Quick Start</a> •
-  <a href="#-why-superlocalemory">Why This?</a> •
+  <a href="#-why-superlocalmemory">Why This?</a> •
   <a href="#-features">Features</a> •
-  <a href="#-vs-alternatives">vs Alternatives</a> •
   <a href="#-documentation">Docs</a> •
   <a href="https://github.com/varun369/SuperLocalMemoryV2/issues">Issues</a>
 </p>
@@ -45,77 +44,106 @@
 
 ---
 
-## NEW: v2.6 — Security Hardening & Performance
+## What's New in v2.7 — "Your AI Learns You"
 
-> **SuperLocalMemory is now production-hardened with trust enforcement, rate limiting, and accelerated graph building.**
+**SuperLocalMemory now learns your patterns, adapts to your workflow, and personalizes recall — all 100% locally on your machine.** No cloud. No LLM. Your behavioral data never leaves your device.
 
-| What's New in v2.6 | Why It Matters |
-|---------------------|----------------|
-| **Trust Enforcement** | Agents with trust below 0.3 are blocked from write/delete — Bayesian scoring now actively protects your memory. |
-| **Profile Isolation** | Memory profiles are fully sandboxed — no cross-profile data leakage. |
-| **Rate Limiting** | Protects against memory flooding and spam from misbehaving agents. |
-| **SSRF Protection** | Webhook dispatcher validates URLs to prevent server-side request forgery. |
-| **HNSW-Accelerated Graphs** | Knowledge graph edge building uses HNSW index for faster construction at scale. |
-| **Hybrid Search Engine** | Combined semantic + FTS5 + graph retrieval for maximum accuracy. |
+### Adaptive Learning System
 
-**v2.5 highlights (included):** Real-time event stream, WAL-mode concurrent writes, agent tracking, memory provenance, 28 API endpoints.
+Your memory system evolves with you through three learning layers:
 
-**Upgrade:** `npm install -g superlocalmemory@latest`
+| Layer | What It Learns | How |
+|-------|---------------|-----|
+| **Tech Preferences** | "You prefer FastAPI over Django" (83% confidence) | Cross-project frequency analysis with Bayesian scoring |
+| **Project Context** | Detects your active project from 4 signals | Path analysis, tags, profile, content clustering |
+| **Workflow Patterns** | "You typically: docs → architecture → code → test" | Time-weighted sliding-window sequence mining |
 
-**Dashboard:** `python3 ~/.claude-memory/ui_server.py` then open `http://localhost:8765`
+### Three-Phase Adaptive Ranking
 
-[Interactive Architecture Diagram](https://superlocalmemory.com/architecture.html) | [Architecture Doc](docs/ARCHITECTURE-V2.5.md) | [Full Changelog](CHANGELOG.md)
+Recall results get smarter over time — automatically:
 
----
+1. **Phase 1 (Baseline):** Standard search — same as v2.6
+2. **Phase 2 (Rule-Based):** After ~20 feedback signals — boosts results matching your preferences
+3. **Phase 3 (ML Ranking):** After ~200 signals — LightGBM LambdaRank re-ranks with 9 personalized features
 
-## NEW: Framework Integrations (v2.5.1)
+### Privacy by Design — GDPR Compliant
 
-Use SuperLocalMemory as a memory backend in your LangChain and LlamaIndex applications — 100% local, zero cloud.
+| Concern | SuperLocalMemory v2.7 | Cloud-Based Alternatives |
+|---------|----------------------|--------------------------|
+| **Where is learning data?** | `~/.claude-memory/learning.db` on YOUR machine | Their servers, their terms |
+| **Who processes your behavior?** | Local gradient boosting (no LLM, no GPU) | Cloud LLMs process your data |
+| **Right to erasure (GDPR Art. 17)?** | `slm learning reset` — one command, instant | Submit a request, wait weeks |
+| **Data portability?** | Copy the SQLite file | Vendor lock-in |
+| **Telemetry?** | Zero. Absolutely none. | Usage analytics, behavior tracking |
 
-### LangChain
+**Your learning data is stored separately from your memories.** Delete `learning.db` and your memories are untouched. Delete `memory.db` and your learning patterns are untouched. Full data sovereignty.
 
-```bash
-pip install langchain-superlocalmemory
-```
+### Research-Backed Architecture
 
-```python
-from langchain_superlocalmemory import SuperLocalMemoryChatMessageHistory
-from langchain_core.runnables.history import RunnableWithMessageHistory
+Every component is grounded in peer-reviewed research, adapted for local-first operation:
 
-history = SuperLocalMemoryChatMessageHistory(session_id="my-session")
-# Messages persist across sessions, stored locally in ~/.claude-memory/memory.db
-```
+| Component | Research Basis |
+|-----------|---------------|
+| Two-stage retrieval pipeline | BM25 → re-ranker (eKNOW 2025) |
+| Adaptive cold-start ranking | Hierarchical meta-learning (LREC 2024) |
+| Time-weighted sequence mining | TSW-PrefixSpan (IEEE 2020) |
+| Bayesian confidence scoring | MACLA (arXiv:2512.18950) |
+| LightGBM LambdaRank | Pairwise ranking (Burges 2010, MO-LightGBM SIGIR 2025) |
+| Privacy-preserving feedback | Zero-communication design — stronger than differential privacy (ADPMF, IPM 2024) |
 
-### LlamaIndex
+### New MCP Tools
 
-```bash
-pip install llama-index-storage-chat-store-superlocalmemory
-```
+| Tool | Purpose |
+|------|---------|
+| `memory_used` | Tell the AI which recalled memories were useful — trains the ranking model |
+| `get_learned_patterns` | See what the system has learned about your preferences |
+| `correct_pattern` | Fix a wrong pattern — your correction overrides with maximum confidence |
 
-```python
-from llama_index.storage.chat_store.superlocalmemory import SuperLocalMemoryChatStore
-from llama_index.core.memory import ChatMemoryBuffer
+### New CLI Commands
 
-chat_store = SuperLocalMemoryChatStore()
-memory = ChatMemoryBuffer.from_defaults(chat_store=chat_store, chat_store_key="user-1")
+```bash
+slm useful 42 87           # Mark memories as useful (ranking feedback)
+slm patterns list           # See learned tech preferences
+slm learning status         # Learning system diagnostics
+slm learning reset          # Delete all behavioral data (memories preserved)
+slm engagement              # Local engagement health metrics
 ```
 
-[LangChain Guide](https://github.com/varun369/SuperLocalMemoryV2/wiki/LangChain-Integration) | [LlamaIndex Guide](https://github.com/varun369/SuperLocalMemoryV2/wiki/LlamaIndex-Integration)
+**Upgrade:** `npm install -g superlocalmemory@latest` — Learning dependencies install automatically.
+
+[Learning System Guide →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Learning-System) | [Upgrade Guide →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Upgrading-to-v2.7) | [Full Changelog](CHANGELOG.md)
 
 ---
 
-## Install in One Command
+<details>
+<summary><strong>Previous: v2.6.5 — Interactive Knowledge Graph</strong></summary>
 
-```bash
-npm install -g superlocalmemory
-```
+- Fully interactive visualization with zoom, pan, click-to-explore (Cytoscape.js)
+- 6 layout algorithms, smart cluster filtering, 10,000+ node performance
+- Mobile & accessibility support: touch gestures, keyboard nav, screen reader
 
-Or clone manually:
-```bash
-git clone https://github.com/varun369/SuperLocalMemoryV2.git && cd SuperLocalMemoryV2 && ./install.sh
-```
+</details>
 
-Both methods auto-detect and configure **17+ IDEs and AI tools** — Cursor, VS Code/Copilot, Codex, Claude, Windsurf, Gemini CLI, JetBrains, and more.
+<details>
+<summary><strong>Previous: v2.6 — Security & Scale</strong></summary>
+
+## What's New in v2.6
+
+SuperLocalMemory is now **production-hardened** with security, performance, and scale improvements:
+
+- **Trust Enforcement** — Bayesian scoring actively protects your memory. Agents with trust below 0.3 are blocked from write/delete operations.
+- **Profile Isolation** — Memory profiles fully sandboxed. Zero cross-profile data leakage.
+- **Rate Limiting** — Protects against memory flooding from misbehaving agents.
+- **HNSW-Accelerated Graphs** — Knowledge graph edge building uses HNSW index for faster construction at scale.
+- **Hybrid Search Engine** — Combined semantic + FTS5 + graph retrieval for maximum accuracy.
+
+**v2.5 highlights (included):** Real-time event stream, WAL-mode concurrent writes, agent tracking, memory provenance, 28 API endpoints.
+
+**Upgrade:** `npm install -g superlocalmemory@latest`
+
+[Interactive Architecture Diagram](https://superlocalmemory.com/architecture.html) | [Architecture Doc](docs/ARCHITECTURE-V2.5.md) | [Full Changelog](CHANGELOG.md)
+
+</details>
 
 ---
 
@@ -135,6 +163,8 @@ You: *sighs and explains everything again*
 - Previous decisions
 - Debugging history
 
+---
+
 ## The Solution
 
 ```bash
@@ -155,26 +185,21 @@ superlocalmemoryv2:recall "auth bug"
 
 ## 🚀 Quick Start
 
-### npm (Recommended — All Platforms)
+### Install (One Command)
+
 ```bash
 npm install -g superlocalmemory
 ```
 
-### Mac/Linux (Manual)
+Or clone manually:
 ```bash
-git clone https://github.com/varun369/SuperLocalMemoryV2.git
-cd SuperLocalMemoryV2
-./install.sh
+git clone https://github.com/varun369/SuperLocalMemoryV2.git && cd SuperLocalMemoryV2 && ./install.sh
 ```
 
-### Windows (PowerShell)
-```powershell
-git clone https://github.com/varun369/SuperLocalMemoryV2.git
-cd SuperLocalMemoryV2
-.\install.ps1
-```
+Both methods auto-detect and configure **17+ IDEs and AI tools** — Cursor, VS Code/Copilot, Codex, Claude, Windsurf, Gemini CLI, JetBrains, and more.
 
 ### Verify Installation
+
 ```bash
 superlocalmemoryv2:status
 # ✓ Database: OK (0 memories)
@@ -184,181 +209,107 @@ superlocalmemoryv2:status
 
 **That's it.** No Docker. No API keys. No cloud accounts. No configuration.
 
-### Updating to Latest Version
+### Launch Dashboard
 
-**npm users:**
 ```bash
-# Update to latest version
-npm update -g superlocalmemory
-
-# Or force latest
-npm install -g superlocalmemory@latest
-
-# Install specific version
-npm install -g superlocalmemory@latest
-```
-
-**Manual install users:**
-```bash
-cd SuperLocalMemoryV2
-git pull origin main
-./install.sh  # Mac/Linux
-# or
-.\install.ps1  # Windows
-```
-
-**Your data is safe:** Updates preserve your database and all memories.
-
-### Start the Visualization Dashboard
-
-```bash
-# Launch the interactive web UI
+# Start the interactive web UI
 python3 ~/.claude-memory/ui_server.py
 
 # Opens at http://localhost:8765
-# Features: Timeline view, search explorer, graph visualization
+# Features: Timeline, search, interactive graph, statistics
 ```
 
 ---
 
-## 🎨 Visualization Dashboard
-
-**NEW in v2.2.0:** Interactive web-based dashboard for exploring your memories visually.
-
-### Features
-
-| Feature | Description |
-|---------|-------------|
-| **📈 Timeline View** | See your memories chronologically with importance indicators |
-| **🔍 Search Explorer** | Real-time semantic search with score visualization |
-| **🕸️ Graph Visualization** | Interactive knowledge graph with clusters and relationships |
-| **📊 Statistics Dashboard** | Memory trends, tag clouds, pattern insights |
-| **🎯 Advanced Filters** | Filter by tags, importance, date range, clusters |
-
-### Quick Tour
+## 💡 Why SuperLocalMemory?
 
-```bash
-# 1. Start dashboard
-python ~/.claude-memory/ui_server.py
+### For Developers Who Use AI Daily
 
-# 2. Navigate to http://localhost:8765
+| Scenario | Without Memory | With SuperLocalMemory |
+|----------|---------------|----------------------|
+| New Claude session | Re-explain entire project | `recall "project context"` → instant context |
+| Debugging | "We tried X last week..." starts over | Knowledge graph shows related past fixes |
+| Code preferences | "I prefer React..." every time | Pattern learning knows your style |
+| Multi-project | Context constantly bleeds | Separate profiles per project |
 
-# 3. Explore your memories:
-#    - Timeline: See memories over time
-#    - Search: Find with semantic scoring
-#    - Graph: Visualize relationships
-#    - Stats: Analyze patterns
-```
+### Built on 2026 Research
 
-**[[Complete Dashboard Guide →|Visualization-Dashboard]]**
+Not another simple key-value store. SuperLocalMemory implements **cutting-edge memory architecture** backed by peer-reviewed research:
 
----
+- **PageIndex** (Meta AI) → Hierarchical memory organization
+- **GraphRAG** (Microsoft) → Knowledge graph with auto-clustering
+- **xMemory** (Stanford) → Identity pattern learning
+- **A-RAG** → Multi-level retrieval with context awareness
+- **LambdaRank** (Burges 2010, MO-LightGBM SIGIR 2025) → Adaptive re-ranking *(v2.7)*
+- **TSW-PrefixSpan** (IEEE 2020) → Time-weighted workflow pattern mining *(v2.7)*
+- **MACLA** (arXiv:2512.18950) → Bayesian temporal confidence scoring *(v2.7)*
+- **FCS** (LREC 2024) → Hierarchical cold-start mitigation *(v2.7)*
 
-### New in v2.4.1: Hierarchical Clustering, Community Summaries & Auto-Backup
+**The only open-source implementation combining all eight approaches — entirely locally.**
 
-| Feature | Description |
-|---------|-------------|
-| **Hierarchical Leiden** | Recursive community detection — clusters within clusters up to 3 levels. "Python" → "FastAPI" → "Auth patterns" |
-| **Community Summaries** | TF-IDF structured reports per cluster: key topics, projects, categories at a glance |
-| **MACLA Confidence** | Bayesian Beta-Binomial scoring (arXiv:2512.18950) — calibrated confidence, not raw frequency |
-| **Auto-Backup** | Configurable SQLite backups with retention policies, restore from any backup via CLI |
-| **Profile UI** | Create, switch, delete profiles from the web dashboard — full isolation per context |
-| **Profile Isolation** | All API endpoints (graph, clusters, patterns, timeline) scoped to active profile |
+[See research citations →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Research-Foundations)
 
 ---
 
-## 🔍 Advanced Search
-
-SuperLocalMemory V2.2.0 implements **hybrid search** combining multiple strategies for maximum accuracy.
-
-### Search Strategies
-
-| Strategy | Method | Best For |
-|----------|--------|----------|
-| **Semantic Search** | TF-IDF vectors + cosine similarity | Conceptual queries ("authentication patterns") |
-| **Full-Text Search** | SQLite FTS5 with ranking | Exact phrases ("JWT tokens expire") |
-| **Graph-Enhanced** | Knowledge graph traversal | Related concepts ("show auth-related") |
-| **Hybrid Mode** | All three combined | General queries (default) |
-
-### Search Examples
-
-```bash
-# Semantic: finds conceptually similar
-slm recall "security best practices"
-# Matches: "JWT implementation", "OAuth flow", "CSRF protection"
+## ✨ Features
 
-# Exact: finds literal text
-slm recall "PostgreSQL 15"
-# Matches: exactly "PostgreSQL 15"
+### Multi-Layer Memory Architecture
 
-# Graph: finds related via clusters
-slm recall "authentication" --use-graph
-# Matches: JWT, OAuth, sessions (via "Auth & Security" cluster)
+**[View Interactive Architecture Diagram](https://superlocalmemory.com/architecture.html)** — Click any layer for details, research references, and file paths.
 
-# Hybrid: best of all worlds (default)
-slm recall "API design patterns"
-# Combines semantic + exact + graph for optimal results
+```
+┌─────────────────────────────────────────────────────────────┐
+│  Layer 9: VISUALIZATION (v2.2+)                             │
+│  Interactive dashboard: timeline, graph explorer, analytics │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 8: HYBRID SEARCH (v2.2+)                             │
+│  Combines: Semantic + FTS5 + Graph traversal                │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 7: UNIVERSAL ACCESS                                  │
+│  MCP + Skills + CLI (works everywhere)                      │
+│  17+ IDEs with single database                              │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 6: MCP INTEGRATION                                   │
+│  Model Context Protocol: 12 tools, 6 resources, 2 prompts   │
+│  Auto-configured for Cursor, Windsurf, Claude               │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 5½: ADAPTIVE LEARNING (v2.7 — NEW)                   │
+│  Three-layer learning: tech prefs + project context + flow  │
+│  LightGBM LambdaRank re-ranking (fully local, no cloud)    │
+│  Research: eKNOW 2025, MACLA, TSW-PrefixSpan, LREC 2024    │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 5: SKILLS LAYER                                      │
+│  7 universal slash-commands for AI assistants               │
+│  Compatible with Claude Code, Continue, Cody                │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 4: PATTERN LEARNING + MACLA                          │
+│  Bayesian confidence scoring (arXiv:2512.18950)             │
+│  "You prefer React over Vue" (73% confidence)               │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 3: KNOWLEDGE GRAPH + HIERARCHICAL CLUSTERING         │
+│  Recursive Leiden algorithm: "Python" → "FastAPI" → "Auth" │
+│  Community summaries with TF-IDF structured reports         │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 2: HIERARCHICAL INDEX                                │
+│  Tree structure for fast navigation                         │
+│  O(log n) lookups instead of O(n) scans                     │
+├─────────────────────────────────────────────────────────────┤
+│  Layer 1: RAW STORAGE                                       │
+│  SQLite + Full-text search + TF-IDF vectors                 │
+│  Compression: 60-96% space savings                          │
+└─────────────────────────────────────────────────────────────┘
 ```
 
-### Measured Search Latency
-
-| Database Size | Median | P95 | P99 |
-|---------------|--------|-----|-----|
-| 100 memories | **10.6ms** | 14.9ms | 15.8ms |
-| 500 memories | **65.2ms** | 101.7ms | 112.5ms |
-| 1,000 memories | **124.3ms** | 190.1ms | 219.5ms |
-
-For typical personal databases (under 500 memories), search returns faster than you blink. [Full benchmarks →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Performance-Benchmarks)
-
----
-
-## ⚡ Measured Performance
-
-All numbers measured on real hardware (Apple M4 Pro, 24GB RAM). No estimates — real benchmarks.
-
-### Search Speed
-
-| Database Size | Median Latency | P95 Latency |
-|---------------|----------------|-------------|
-| 100 memories | **10.6ms** | 14.9ms |
-| 500 memories | **65.2ms** | 101.7ms |
-| 1,000 memories | **124.3ms** | 190.1ms |
-
-For typical personal use (under 500 memories), search results return faster than you blink.
-
-### Concurrent Writes — Zero Errors
-
-| Scenario | Writes/sec | Errors |
-|----------|------------|--------|
-| 1 AI tool writing | **204/sec** | 0 |
-| 2 AI tools simultaneously | **220/sec** | 0 |
-| 5 AI tools simultaneously | **130/sec** | 0 |
-| 10 AI tools simultaneously | **25/sec** | 0 |
-
-WAL mode + serialized write queue = zero "database is locked" errors, ever.
-
-### Storage
-
-10,000 memories = **13.6 MB** on disk (~1.9 KB per memory). Your entire AI memory history takes less space than a photo.
-
-### Trust Defense
-
-Bayesian trust scoring achieves **perfect separation** (trust gap = 1.0) between honest and malicious agents. Detects "sleeper" attacks with 74.7% trust drop. Zero false positives.
-
-### Graph Construction
-
-| Memories | Build Time |
-|----------|-----------|
-| 100 | 0.28s |
-| 1,000 | 10.6s |
-
-Leiden clustering discovers 6-7 natural topic communities automatically.
-
-> **Graph Scaling:** Knowledge graph features work best with up to 10,000 memories. For larger databases, the system uses intelligent sampling (most recent + highest importance memories) for graph construction. Core search and memory storage have no upper limit.
-
-> **LoCoMo benchmark results coming soon** — evaluation against the standardized [LoCoMo](https://snap-research.github.io/locomo/) long-conversation memory benchmark (Snap Research, ACL 2024).
+### Key Capabilities
 
-[Full benchmark details →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Performance-Benchmarks)
+- **[Adaptive Learning System](https://github.com/varun369/SuperLocalMemoryV2/wiki/Learning-System)** — Learns your tech preferences, workflow patterns, and project context. Personalizes recall ranking using local ML (LightGBM). Zero cloud dependency. *New in v2.7*
+- **[Knowledge Graphs](https://github.com/varun369/SuperLocalMemoryV2/wiki/Knowledge-Graph)** — Automatic relationship discovery. Interactive visualization with zoom, pan, click.
+- **[Pattern Learning](https://github.com/varun369/SuperLocalMemoryV2/wiki/Pattern-Learning)** — Learns your coding preferences and style automatically.
+- **[Multi-Profile Support](https://github.com/varun369/SuperLocalMemoryV2/wiki/Using-Memory-Profiles)** — Isolated contexts for work, personal, clients. Zero context bleeding.
+- **[Hybrid Search](https://github.com/varun369/SuperLocalMemoryV2/wiki/Advanced-Search)** — Semantic + FTS5 + Graph retrieval combined for maximum accuracy.
+- **[Visualization Dashboard](https://github.com/varun369/SuperLocalMemoryV2/wiki/Visualization-Dashboard)** — Web UI for timeline, search, graph exploration, analytics.
+- **[Framework Integrations](docs/FRAMEWORK-INTEGRATIONS.md)** — Use with LangChain and LlamaIndex applications.
+- **[Real-Time Events](https://github.com/varun369/SuperLocalMemoryV2/wiki/Real-Time-Event-System)** — Live notifications via SSE/WebSocket/Webhooks when memories change.
 
 ---
 
@@ -381,79 +332,29 @@ Leiden clustering discovers 6-7 natural topic communities automatically.
 | **Gemini CLI** | ✅ MCP + Skills | Native MCP + skills |
 | **JetBrains IDEs** | ✅ MCP | Via AI Assistant settings |
 | **Zed Editor** | ✅ MCP | Native MCP tools |
-| **OpenCode** | ✅ MCP | Native MCP tools |
-| **Perplexity** | ✅ MCP | Native MCP tools |
-| **Antigravity** | ✅ MCP + Skills | Native MCP tools |
-| **ChatGPT** | ✅ MCP Connector | `search()` + `fetch()` via HTTP tunnel |
 | **Aider** | ✅ Smart Wrapper | `aider-smart` with context |
 | **Any Terminal** | ✅ Universal CLI | `slm remember "content"` |
 
 ### Three Ways to Access
 
-1. **MCP (Model Context Protocol)** - Auto-configured for Cursor, Windsurf, Claude Desktop
+1. **MCP (Model Context Protocol)** — Auto-configured for Cursor, Windsurf, Claude Desktop
    - AI assistants get natural access to your memory
    - No manual commands needed
    - "Remember that we use FastAPI" just works
 
-2. **Skills & Commands** - For Claude Code, Continue.dev, Cody
+2. **Skills & Commands** — For Claude Code, Continue.dev, Cody
    - `/superlocalmemoryv2:remember` in Claude Code
    - `/slm-remember` in Continue.dev and Cody
    - Familiar slash command interface
 
-3. **Universal CLI** - Works in any terminal or script
+3. **Universal CLI** — Works in any terminal or script
    - `slm remember "content"` - Simple, clean syntax
    - `slm recall "query"` - Search from anywhere
    - `aider-smart` - Aider with auto-context injection
 
 **All three methods use the SAME local database.** No data duplication, no conflicts.
 
-### Auto-Detection
-
-Installation automatically detects and configures:
-- Existing IDEs (Cursor, Windsurf, VS Code)
-- Installed tools (Aider, Continue, Cody)
-- Shell environment (bash, zsh)
-
-**Zero manual configuration required.** It just works.
-
-### Manual Setup for Other Apps
-
-Want to use SuperLocalMemory in ChatGPT, Perplexity, Zed, or other MCP-compatible tools?
-
-**📘 Complete setup guide:** [docs/MCP-MANUAL-SETUP.md](docs/MCP-MANUAL-SETUP.md)
-
-Covers:
-- ChatGPT Desktop - Add via Settings → MCP
-- Perplexity - Configure via app settings
-- Zed Editor - JSON configuration
-- Cody - VS Code/JetBrains setup
-- Custom MCP clients - Python/HTTP integration
-
-All tools connect to the **same local database** - no data duplication.
-
----
-
-## 💡 Why SuperLocalMemory?
-
-### For Developers Who Use AI Daily
-
-| Scenario | Without Memory | With SuperLocalMemory |
-|----------|---------------|----------------------|
-| New Claude session | Re-explain entire project | `recall "project context"` → instant context |
-| Debugging | "We tried X last week..." starts over | Knowledge graph shows related past fixes |
-| Code preferences | "I prefer React..." every time | Pattern learning knows your style |
-| Multi-project | Context constantly bleeds | Separate profiles per project |
-
-### Built on 2026 Research
-
-Not another simple key-value store. SuperLocalMemory implements **cutting-edge memory architecture**:
-
-- **PageIndex** (Meta AI) → Hierarchical memory organization
-- **GraphRAG** (Microsoft) → Knowledge graph with auto-clustering
-- **xMemory** (Stanford) → Identity pattern learning
-- **A-RAG** → Multi-level retrieval with context awareness
-
-**The only open-source implementation combining all four approaches.**
+[Complete setup guide for all tools →](docs/MCP-MANUAL-SETUP.md)
 
 ---
 
@@ -470,150 +371,72 @@ Not another simple key-value store. SuperLocalMemory implements **cutting-edge m
 | **Letta/MemGPT** | Self-hosted (complex) | TBD | Requires significant setup |
 | **SuperLocalMemory V2** | **Unlimited** | **$0 forever** | **Nothing.** |
 
-### Feature Comparison (What Actually Matters)
+### What Actually Matters
 
 | Feature | Mem0 | Zep | Khoj | Letta | **SuperLocalMemory V2** |
 |---------|------|-----|------|-------|------------------------|
 | **Works in Cursor** | Cloud Only | ❌ | ❌ | ❌ | ✅ **Local** |
 | **Works in Windsurf** | Cloud Only | ❌ | ❌ | ❌ | ✅ **Local** |
 | **Works in VS Code** | 3rd Party | ❌ | Partial | ❌ | ✅ **Native** |
-| **Works in Claude** | ❌ | ❌ | ❌ | ❌ | ✅ |
-| **Works with Aider** | ❌ | ❌ | ❌ | ❌ | ✅ |
 | **Universal CLI** | ❌ | ❌ | ❌ | ❌ | ✅ |
-| **7-Layer Universal Architecture** | ❌ | ❌ | ❌ | ❌ | ✅ |
+| **Multi-Layer Architecture** | ❌ | ❌ | ❌ | ❌ | ✅ |
 | **Pattern Learning** | ❌ | ❌ | ❌ | ❌ | ✅ |
-| **Multi-Profile Support** | ❌ | ❌ | ❌ | Partial | ✅ |
+| **Adaptive ML Ranking** | Cloud LLM | ❌ | ❌ | ❌ | ✅ **Local ML** |
 | **Knowledge Graphs** | ✅ | ✅ | ❌ | ❌ | ✅ |
 | **100% Local** | ❌ | ❌ | Partial | Partial | ✅ |
+| **GDPR by Design** | ❌ | ❌ | ❌ | ❌ | ✅ |
 | **Zero Setup** | ❌ | ❌ | ❌ | ❌ | ✅ |
-| **Progressive Compression** | ❌ | ❌ | ❌ | ❌ | ✅ |
 | **Completely Free** | Limited | Limited | Partial | ✅ | ✅ |
 
 **SuperLocalMemory V2 is the ONLY solution that:**
+- ✅ **Learns and adapts** locally — no cloud LLM needed for personalization
 - ✅ Works across 17+ IDEs and CLI tools
 - ✅ Remains 100% local (no cloud dependencies)
+- ✅ GDPR Article 17 compliant — one-command data erasure
 - ✅ Completely free with unlimited memories
 
 [See full competitive analysis →](docs/COMPETITIVE-ANALYSIS.md)
 
 ---
 
-## ✨ Features
-
-### Multi-Layer Memory Architecture
-
-**[View Interactive Architecture Diagram](https://superlocalmemory.com/architecture.html)** — Click any layer for details, research references, and file paths.
-
-```
-┌─────────────────────────────────────────────────────────────┐
-│  Layer 9: VISUALIZATION (NEW v2.2.0)                        │
-│  Interactive dashboard: timeline, search, graph explorer    │
-│  Real-time analytics and visual insights                    │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 8: HYBRID SEARCH (NEW v2.2.0)                        │
-│  Combines: Semantic + FTS5 + Graph traversal                │
-│  80ms response time with maximum accuracy                   │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 7: UNIVERSAL ACCESS                                  │
-│  MCP + Skills + CLI (works everywhere)                      │
-│  17+ IDEs with single database                              │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 6: MCP INTEGRATION                                   │
-│  Model Context Protocol: 6 tools, 4 resources, 2 prompts    │
-│  Auto-configured for Cursor, Windsurf, Claude               │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 5: SKILLS LAYER                                      │
-│  6 universal slash-commands for AI assistants               │
-│  Compatible with Claude Code, Continue, Cody                │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 4: PATTERN LEARNING + MACLA (v2.4.0)                  │
-│  Bayesian Beta-Binomial confidence (arXiv:2512.18950)       │
-│  "You prefer React over Vue" (73% confidence)               │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 3: KNOWLEDGE GRAPH + HIERARCHICAL LEIDEN (v2.4.1)    │
-│  Recursive clustering: "Python" → "FastAPI" → "Auth"        │
-│  Community summaries + TF-IDF structured reports            │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 2: HIERARCHICAL INDEX                                │
-│  Tree structure for fast navigation                         │
-│  O(log n) lookups instead of O(n) scans                     │
-├─────────────────────────────────────────────────────────────┤
-│  Layer 1: RAW STORAGE                                       │
-│  SQLite + Full-text search + TF-IDF vectors                 │
-│  Compression: 60-96% space savings                          │
-└─────────────────────────────────────────────────────────────┘
-```
-
-### Knowledge Graph (It's Magic)
-
-```bash
-# Build the graph from your memories
-python ~/.claude-memory/graph_engine.py build
-
-# Output:
-# ✓ Processed 47 memories
-# ✓ Created 12 clusters:
-#   - "Authentication & Tokens" (8 memories)
-#   - "Performance Optimization" (6 memories)
-#   - "React Components" (11 memories)
-#   - "Database Queries" (5 memories)
-#   ...
-```
+## ⚡ Measured Performance
 
-**The graph automatically discovers relationships.** Ask "what relates to auth?" and get JWT, session management, token refresh—even if you never tagged them together.
+All numbers measured on real hardware (Apple M4 Pro, 24GB RAM). No estimates — real benchmarks.
 
-### Pattern Learning (It Knows You)
+### Search Speed
 
-```bash
-# Learn patterns from your memories
-python ~/.claude-memory/pattern_learner.py update
-
-# Get your coding identity
-python ~/.claude-memory/pattern_learner.py context 0.5
-
-# Output:
-# Your Coding Identity:
-# - Framework preference: React (73% confidence)
-# - Style: Performance over readability (58% confidence)
-# - Testing: Jest + React Testing Library (65% confidence)
-# - API style: REST over GraphQL (81% confidence)
-```
+| Database Size | Median Latency | P95 Latency |
+|---------------|----------------|-------------|
+| 100 memories | **10.6ms** | 14.9ms |
+| 500 memories | **65.2ms** | 101.7ms |
+| 1,000 memories | **124.3ms** | 190.1ms |
 
-**Your AI assistant can now match your preferences automatically.**
+For typical personal use (under 500 memories), search results return faster than you blink.
 
-**MACLA Confidence Scoring (v2.4.0):** Confidence uses a Bayesian Beta-Binomial posterior (Forouzandeh et al., [arXiv:2512.18950](https://arxiv.org/abs/2512.18950)). Pattern-specific priors, log-scaled competition, recency bonus. Range: 0.0–0.95 (hard cap prevents overconfidence).
+### Concurrent Writes — Zero Errors
 
-### Multi-Profile Support
+| Scenario | Writes/sec | Errors |
+|----------|------------|--------|
+| 1 AI tool writing | **204/sec** | 0 |
+| 2 AI tools simultaneously | **220/sec** | 0 |
+| 5 AI tools simultaneously | **130/sec** | 0 |
 
-```bash
-# Work profile
-superlocalmemoryv2:profile create work --description "Day job"
-superlocalmemoryv2:profile switch work
+WAL mode + serialized write queue = zero "database is locked" errors, ever.
 
-# Personal projects
-superlocalmemoryv2:profile create personal
-superlocalmemoryv2:profile switch personal
+### Storage
 
-# Client projects (completely isolated)
-superlocalmemoryv2:profile create client-acme
-```
+10,000 memories = **13.6 MB** on disk (~1.9 KB per memory). Your entire AI memory history takes less space than a photo.
 
-**Each profile has isolated memories, graphs, and patterns.** No context bleeding.
+### Graph Construction
 
----
+| Memories | Build Time |
+|----------|-----------|
+| 100 | 0.28s |
+| 1,000 | 10.6s |
 
-## 📖 Documentation
+Leiden clustering discovers 6-7 natural topic communities automatically.
 
-| Guide | Description |
-|-------|-------------|
-| [Quick Start](https://github.com/varun369/SuperLocalMemoryV2/wiki/Quick-Start-Tutorial) | Get running in 5 minutes |
-| [Installation](https://github.com/varun369/SuperLocalMemoryV2/wiki/Installation) | Detailed setup instructions |
-| [Visualization Dashboard](https://github.com/varun369/SuperLocalMemoryV2/wiki/Visualization-Dashboard) | Interactive web UI guide (NEW v2.2.0) |
-| [CLI Reference](docs/CLI-COMMANDS-REFERENCE.md) | All commands explained |
-| [Knowledge Graph](docs/GRAPH-ENGINE.md) | How clustering works |
-| [Pattern Learning](docs/PATTERN-LEARNING.md) | Identity extraction |
-| [Profiles Guide](docs/PROFILES-GUIDE.md) | Multi-context management |
-| [API Reference](docs/API-REFERENCE.md) | Python API documentation |
+[Full benchmark details →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Performance-Benchmarks)
 
 ---
 
@@ -632,40 +455,33 @@ superlocalmemoryv2:profile create <name>                 # New profile
 superlocalmemoryv2:profile switch <name>                 # Switch context
 
 # Knowledge Graph
-python ~/.claude-memory/graph_engine.py build            # Build graph (+ hierarchical + summaries)
+python ~/.claude-memory/graph_engine.py build            # Build graph
 python ~/.claude-memory/graph_engine.py stats            # View clusters
-python ~/.claude-memory/graph_engine.py related --id 5   # Find related
-python ~/.claude-memory/graph_engine.py hierarchical     # Sub-cluster large communities
-python ~/.claude-memory/graph_engine.py summaries        # Generate cluster summaries
 
 # Pattern Learning
 python ~/.claude-memory/pattern_learner.py update        # Learn patterns
 python ~/.claude-memory/pattern_learner.py context 0.5   # Get identity
 
-# Auto-Backup (v2.4.0)
-python ~/.claude-memory/auto_backup.py backup            # Manual backup
-python ~/.claude-memory/auto_backup.py list              # List backups
-python ~/.claude-memory/auto_backup.py status            # Backup status
-
-# Reset (Use with caution!)
-superlocalmemoryv2:reset soft                            # Clear memories
-superlocalmemoryv2:reset hard --confirm                  # Nuclear option
+# Visualization Dashboard
+python ~/.claude-memory/ui_server.py                     # Launch web UI
 ```
 
----
+[Complete CLI reference →](docs/CLI-COMMANDS-REFERENCE.md)
 
-## 📊 Performance at a Glance
+---
 
-| Metric | Measured Result |
-|--------|----------------|
-| **Search latency** | **10.6ms** median (100 memories) |
-| **Concurrent writes** | **220/sec** with 2 agents, zero errors |
-| **Storage** | **1.9 KB** per memory at scale (13.6 MB for 10K) |
-| **Trust defense** | **1.0** trust gap (perfect separation) |
-| **Graph build** | **0.28s** for 100 memories |
-| **Search quality** | **MRR 0.90** (first result correct 9/10 times) |
+## 📖 Documentation
 
-[Full benchmark details →](https://github.com/varun369/SuperLocalMemoryV2/wiki/Performance-Benchmarks)
+| Guide | Description |
+|-------|-------------|
+| [Quick Start](https://github.com/varun369/SuperLocalMemoryV2/wiki/Quick-Start-Tutorial) | Get running in 5 minutes |
+| [Installation](https://github.com/varun369/SuperLocalMemoryV2/wiki/Installation) | Detailed setup instructions |
+| [Visualization Dashboard](https://github.com/varun369/SuperLocalMemoryV2/wiki/Visualization-Dashboard) | Interactive web UI guide |
+| [Interactive Graph](https://github.com/varun369/SuperLocalMemoryV2/wiki/Using-Interactive-Graph) | Graph exploration guide (NEW v2.6.5) |
+| [Framework Integrations](docs/FRAMEWORK-INTEGRATIONS.md) | LangChain & LlamaIndex setup |
+| [Knowledge Graph](docs/GRAPH-ENGINE.md) | How clustering works |
+| [Pattern Learning](docs/PATTERN-LEARNING.md) | Identity extraction |
+| [API Reference](docs/API-REFERENCE.md) | Python API documentation |
 
 ---
 
@@ -675,9 +491,8 @@ We welcome contributions! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
 
 **Areas for contribution:**
 - Additional pattern categories
-- Graph visualization UI
-- Integration with more AI assistants
 - Performance optimizations
+- Integration with more AI assistants
 - Documentation improvements
 
 ---