claude-faf-mcp 2.6.4 → 2.6.6

package/CHANGELOG.md

@@ -5,6 +5,24 @@ All notable changes to claude-faf-mcp will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [2.6.6] - 2025-10-25
+
+### Changed
+- Redesigned README for professional clarity and credibility
+- Removed decorative emoji and formatting from meta-content
+- Emphasized "persistent project context" in official registry listing
+- Clarified format-driven architecture (format-first, not tools-first)
+
+### Added
+- "Scoring System Experience" section with product screenshot
+- Homebrew installation option (brew install Wolfe-Jam/tap/claude-faf-mcp)
+- Strengthened positioning: "first and only persistent project context server"
+
+### Documentation
+- Format-driven architecture now leads value proposition
+- Distinguished .faf from tools.md and CLAUDE.md
+- Professional, noise-free README structure
+
 ## [2.6.3] - 2025-10-20
 
 ### Fixed

package/CLAUDE.md

@@ -1,289 +1,30 @@
-# 🧡⚡️claude-faf-mcp v2.6.0 🏆 Claude Desktop MCP Server
-**Project DNA ✨ for AI-Context, On-Demand**
+# 🏎️ CLAUDE.md - claude-faf-mcp Persistent Context & Intelligence
 
-## Project Overview
-**Package:** claude-faf-mcp
-**Version:** 2.6.0
-**Stack:** TypeScript 100%, Node.js 18+
-**License:** MIT (FREE FOREVER)
-**Status:** Published on npm, 800+ weekly downloads
+## PROJECT STATE: GOOD 🚀
+**Current Position:** Provide FAF context tools directly in Claude Desktop via MCP
+**Tyre Compound:** ULTRASOFT C5 (Maximum Performance)
 
-## What is claude-faf-mcp?
-Model Context Protocol (MCP) server that brings .faf Project DNA ✨ directly into Claude Desktop for AI-Context, On-Demand. 33+ tools for creating, scoring, syncing, and managing AI context without leaving your conversation.
-
-## MCP Architecture
-**Protocol:** Model Context Protocol (Anthropic)
-**Transport:** stdio (standard input/output)
-**SDK:** @modelcontextprotocol/sdk ^1.0.0
-**Tools:** 33 registered MCP tools
-
-## Installation
-Add to Claude Desktop config (`~/Library/Application Support/Claude/claude_desktop_config.json`):
-
-```json
-{
-  "mcpServers": {
-    "faf": {
-      "command": "npx",
-      "args": ["-y", "claude-faf-mcp"]
-    }
-  }
-}
-```
-
-## 33 MCP Tools
-
-### 🏎️ Power Tools
-- `faf_init` - Create .faf from project
-- `faf_auto` - Zero to Championship (auto-detect)
-- `faf_score` - Rate AI readiness (0-100%)
-- `faf_enhance` - AI-powered improvements
-
-### 😽 TURBO-CAT Tools
-- `faf_formats` - 153 format discovery
-- `faf_stacks` - Technology signatures
-- `faf_vibe` - Platform detection
-
-### 🧬 DNA & Evolution
-- `faf_dna` - Show evolution journey
-- `faf_log` - Complete history
-- `faf_update` - Save checkpoint
-- `faf_recover` - Disaster recovery
-- `faf_auth` - Birth certificate
-
-### 🏆 Scoring & Trust
-- `faf_score` - Context quality rating
-- `faf_trust` - Unified dashboard
-- `faf_credit` - Technical credit
-- `faf_check` - Validation
-- `faf_validate` - Structure check
-- `faf_audit` - Freshness check
-
-### 🔗 Sync & Mirror
-- `faf_bi_sync` - .faf ↔ CLAUDE.md (8ms!)
-- `faf_sync` - Dependency updates
-- `faf_tsa` - Inspect deps
-
-### 📝 Content Management
-- `faf_show` - Display content
-- `faf_edit` - Interactive editor
-- `faf_convert` - Format conversion
-- `faf_share` - Secure sharing
-
-### 🔍 Discovery & Analysis
-- `faf_search` - Content search
-- `faf_doctor` - Diagnose issues
-- `faf_verify` - AI compatibility
-
-### 🏁 Family & Integrations
-- `faf_fam` - View integrations (6 total)
-- `faf_fam_show` - Integration details
-
-### 📊 Information
-- `faf_status` - Quick health check
-- `faf_version` - Version info
-- `faf_welcome` - Onboarding guide
-
-## Key Features
-- **Zero Setup** - npx auto-installs, no config needed
-- **Real-time Context** - Tools execute in your project directory
-- **Championship Scoring** - 🏆 🥇 🥈 🥉 🟢 🟡 🔴 🤍
-- **Network Effects** - Works with 6 integrations (React, Next, Svelte, TS, Vite, n8n)
-- **Bi-directional Sync** - 8ms .faf ↔ CLAUDE.md mirroring
-
-## Architecture
-```
-claude-faf-mcp/
-├── src/
-│   ├── index.ts        # MCP server entry
-│   ├── cli.ts          # CLI wrapper
-│   ├── server.ts       # MCP implementation
-│   ├── tools/          # 33 MCP tool definitions
-│   └── handlers/       # Tool execution handlers
-├── dist/               # Compiled output
-├── assets/             # Icons, branding
-└── tests/              # Jest test suites
-```
-
-## Tool Definition Pattern
-Each MCP tool follows this structure:
-
-```typescript
-{
-  name: "faf_[command]",
-  description: "User-facing description for Claude",
-  inputSchema: {
-    type: "object",
-    properties: {
-      path: { type: "string", description: "Project path" },
-      // ... other params
-    }
-  }
-}
-```
-
-## Handler Pattern
-Handlers execute faf-cli commands via child_process:
-
-```typescript
-async function handleFafCommand(
-  command: string,
-  args: string[] = [],
-  cwd?: string
-): Promise<MCPResult> {
-  const { stdout, stderr } = await execCommand(
-    `faf ${command} ${args.join(' ')}`,
-    { cwd }
-  );
-  return { content: [{ type: "text", text: stdout }] };
-}
-```
-
-## Development
-```bash
-# Clone
-git clone https://github.com/Wolfe-Jam/claude-faf-mcp.git
-cd claude-faf-mcp
-
-# Install
-npm install
-
-# Build
-npm run build         # TypeScript → dist/
-
-# Test
-npm test             # Jest test suite
-npm run test:mcp     # MCP inspector
-
-# Dev modes
-npm run dev:stdio    # stdio transport (default)
-npm run dev:http     # HTTP-SSE transport
-npm run debug        # Node debugger
-```
-
-## Testing with MCP Inspector
-```bash
-# Start inspector
-npm run test:mcp
-
-# Or manually
-mcp-inspector stdio ts-node src/cli.ts
-```
-
-## Publishing
-```bash
-# Pre-publish
-npm run build        # Compile TypeScript
-npm test             # Run test suite
-
-# Publish
-npm publish          # Runs prepublishOnly automatically
-```
-
-## Dependencies
-**Production:** 1 only
-- `@modelcontextprotocol/sdk` ^1.0.0
-
-**Dev:** TypeScript 5.3.0, Jest 29.7.0, ts-node, eslint
-
-## Network Effects Strategy
-**Keywords (82 total):**
-- Core: `claude`, `mcp`, `model-context-protocol`, `claude-desktop`
-- Integrations: `chatgpt`, `gemini`, `cursor`, `codex`, `windsurf`, `warp`
-- Context: `ai-context`, `context-management`, `context-mirroring`
-- Features: `championship-scoring`, `bi-sync`, `zero-config`
-
-## Integration with faf-cli
-This MCP server wraps faf-cli commands:
-- **CLI required:** faf-cli must be installed globally or locally
-- **Execution:** Tools spawn `faf [command]` as child processes
-- **Output:** Captures stdout/stderr and returns to Claude
-- **Context:** Executes in user's project directory
-
-## Performance
-- **Tool Registration:** <100ms on startup
-- **Tool Execution:** Inherits faf-cli performance (<50ms)
-- **Bi-sync:** 8ms typical (via faf-cli)
-- **Zero overhead:** Direct CLI passthrough
-
-## Real Stats (Verified)
-- ✅ 800+ weekly downloads
-- ✅ 33 MCP tools
-- ✅ 100% TypeScript strict mode
-- ✅ Works with Claude Desktop
-- ✅ Zero-config installation
-
-## Claude Desktop Integration
-**Location:** `~/Library/Application Support/Claude/claude_desktop_config.json`
-
-**Restart Required:** After config changes, restart Claude Desktop
-
-**Debug Logs:**
-- macOS: `~/Library/Logs/Claude/mcp*.log`
-- Check for connection errors
-
-## Tool Usage in Claude
-Once installed, Claude can use tools naturally:
-
-```
-User: "Initialize a .faf file in my project"
-Claude: [Uses faf_init tool]
-
-User: "What's my AI readiness score?"
-Claude: [Uses faf_score tool]
-
-User: "Show me my project's evolution"
-Claude: [Uses faf_dna tool]
-```
-
-## NO BS ZONE 🛑
-**Trust is everything.**
-- NO fake statistics
-- NO inflated numbers
-- NO guarantees (it's free software)
-- ONLY verified claims
-- ONLY real achievements
-
-## Ecosystem
-- **CLI** - npmjs.com/package/faf-cli (main tool)
-- **MCP Server** (this) - npmjs.com/package/claude-faf-mcp
-- **Chrome Extension** - chromewebstore.google.com (LIVE)
-- **Website** - faf.one
-- **Homebrew** - brew install faf-cli
-
-## Important Context
-- **MCP** = Model Context Protocol (Anthropic standard)
-- **stdio** = Standard input/output (default transport)
-- **Tool** = MCP function Claude can call
-- **Handler** = Server-side function that executes the tool
-- **Orange Smiley MCP** = Our brand/nickname 🧡⚡️
-
-## Git Protocol
-**Professional. Boring. Trusted.**
+---
 
-```
-fix: <what changed>
-feat: <what changed>
-docs: <what changed>
+## 🎨 CORE CONTEXT
 
-- <specific change 1>
-- <specific change 2>
-```
+### Project Identity
+- **Name:** claude-faf-mcp
+- **Stack:** Node.js/TypeScript
+- **Quality:** F1-INSPIRED (Championship Performance)
 
-## Versioning
-- **v2.x** = MCP SDK 1.0 compatibility
-- **Sync with CLI:** MCP versions don't match CLI versions
-- **Independent:** MCP can work with any faf-cli version
+### Technical Architecture
+- **What Building:** Universal FAF MCP Server for Claude Desktop - AI Context Intelligence
+- **Main Language:** TypeScript
 
-## Next Steps
-- Monitor npm download growth
-- Track Claude Desktop adoption
-- Add more MCP-specific tools
-- Improve error handling in stdio
+### 📊 Context Quality Status
+- **Overall Assessment:** Good
+- **Last Updated:** 2025-10-23
 
 ---
 
-**Championship-grade MCP server. Built with 🧡 by wolfejam.**
+**STATUS: BI-SYNC ACTIVE 🔗 - Synchronized with .faf context!**
 
-*Generated: 2025-10-15*
-*Status: v2.6.0 live on npm*
+*Last Sync: 2025-10-23T17:02:56.806Z*
+*Sync Engine: F1-Inspired Software Engineering*
+*🏎️⚡️_championship_sync*

package/README.md

@@ -1,76 +1,97 @@
-# claude-faf-mcp v2.6.4 🏆 1st Persistent Project AI Context MCP Server.
-Now featuring New Skills.md 🎸 integration 
+# claude-faf-mcp
 
-<div align="center">
+## TL;DR
 
-<img src="https://cdn.jsdelivr.net/npm/claude-faf-mcp@latest/assets/icons/faf-icon-64.png" alt="Orange Smiley" width="48" />
+**Problem:** AI needs persistent project context—not just docs or tools, but foundational infrastructure.
 
-**🏆 Anthropic-Approved MCP Server** • **100% error-free TS strict Gold Code Standard** • **Project DNA ✨ for AI-Context, On-Demand.**
+**Solution:** The .faf format is a structured, machine-readable context layer. This MCP server gives Claude 33+ tools to create, score, and improve your project's persistent context through format-driven architecture.
 
-</div>
+**How it works:** Get a score (0-100%) showing how well AI understands your project. Higher scores = AI more in-tune with your codebase. Use tools to improve your score and context quality. Your .faf context persists across sessions.
 
-<div align="center">
+**Install:**
 
-[![NPM Version](https://img.shields.io/npm/v/claude-faf-mcp)](https://www.npmjs.com/package/claude-faf-mcp)
-[![Downloads](https://img.shields.io/npm/dt/claude-faf-mcp)](https://www.npmjs.com/package/claude-faf-mcp)
-[![Weekly Downloads](https://img.shields.io/npm/dw/claude-faf-mcp)](https://www.npmjs.com/package/claude-faf-mcp)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+Via npm:
+```bash
+npm install -g claude-faf-mcp
+```
 
-[![TypeScript](https://img.shields.io/badge/TypeScript-100%25%20Strict-3178C6?logo=typescript)](https://www.typescriptlang.org/)
-[![Node Version](https://img.shields.io/node/v/claude-faf-mcp)](https://nodejs.org)
-[![GitHub Stars](https://img.shields.io/github/stars/Wolfe-Jam/claude-faf-mcp?style=social)](https://github.com/Wolfe-Jam/claude-faf-mcp)
-[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](./CONTRIBUTING.md)
+Via Homebrew:
+```bash
+brew install Wolfe-Jam/tap/claude-faf-mcp
+```
 
-[![Official MCP Registry](https://img.shields.io/badge/Anthropic-Official%20MCP%20Registry-success?logo=github)](https://github.com/modelcontextprotocol/servers)
-[![Production Ready](https://img.shields.io/badge/Status-Production%20Ready-success)](https://wolfe-jam.github.io/claude-faf-mcp/)
-[![Gold Standard](https://img.shields.io/badge/Evaluation-94.4%2F100%20Gold-FFD700)](./docs/evaluations/WJTTC-Comprehensive-Evaluation-2025-10-12.md)
-[![Tests](https://img.shields.io/badge/Tests-730%20C.O.R.E-success)](./docs/evaluations/)
-[![Performance](https://img.shields.io/badge/Performance-%3C11ms-success)](./docs/evaluations/)
+**Configure:** Add to `claude_desktop_config.json`:
+```json
+{
+  "mcpServers": {
+    "claude-faf-mcp": {
+      "command": "claude-faf-mcp"
+    }
+  }
+}
+```
 
-**[Website](https://faf.one)** • **[Docs](https://wolfe-jam.github.io/claude-faf-mcp/)** • **[Community](https://github.com/Wolfe-Jam/claude-faf-mcp/discussions)** • **[Contributing](./CONTRIBUTING.md)**
+**CLI vs MCP clarity**
 
-</div>
+faf-cli (npm) runs on your machine locally in a terminal. claude-faf-mcp (this) runs through Claude Desktop as a tool. Same .faf format, different execution layer. Same persistent context and scoring. Same capabilities (create, score, improve).
+
+Use CLI for raw speed and local development; use MCP for AI-integrated workflows. No feature gaps between them - pick based on your flow.
+
+Website: https://faf.one | GitHub: https://github.com/Wolfe-Jam/claude-faf-mcp | Discussions: https://github.com/Wolfe-Jam/claude-faf-mcp/discussions
 
 ---
 
-## 🏆 Championship Scoring System
+## Official Status
 
-![FAF Championship Score Card](https://cdn.jsdelivr.net/npm/claude-faf-mcp@latest/assets/faf-championship-scorecard.png)
+claude-faf-mcp is officially published in the Anthropic MCP Registry (PR #2759). This is the first and only persistent project context server in the official Anthropic ecosystem.
 
-Track your project's AI-readiness with F1-inspired tiers:
+Registry listing: "MCP server for .faf format. The only persistent project context scoring engine in the Anthropic registry."
 
-```
-🏆 Trophy (100%)    - Championship - Perfect AI|HUMAN balance
-🥇 Gold (99%)       - Gold standard
-🥈 Silver (95-98%)  - Excellence
-🥉 Bronze (85-94%)  - Production ready
-🟢 Green (70-84%)   - Good foundation
-🟡 Yellow (55-69%)  - Getting there
-🔴 Red (0-54%)      - Needs attention
-```
+Published to official Anthropic MCP registry with validation by Anthropic engineering team. Current metrics: 4,100+ total downloads with 462 downloads per week (4.5x growth from 108/week baseline).
 
-**Live output in Claude Desktop:**
+---
 
-```
-🥉 Score: 88/100
-█████████████████████░░░ 88%
-Status: Bronze - Production Ready
+## What is claude-faf-mcp?
 
-Next milestone: 95% 🥈 Silver (7 points to go!)
-```
+An MCP server that brings the .faf format to Claude Desktop for persistent project context. The .faf format (Foundational AI-Context Format) is a structured, machine-readable context layer designed as foundational infrastructure—not tools, not documentation, but format.
+
+**Format-Driven Architecture**
+
+Everything flows through structured format. The .faf file is your project's persistent context layer. It survives across sessions, tools, and AI systems without re-explanation. It works with any MCP client, CLI, workflow automation (n8n, Make, etc.), or AI assistant. It supports any language, framework, or project setup. Optimized for Claude Desktop while maintaining compatibility with any AI model or platform.
+
+Format-driven means the architecture is built on data structure first, not tooling first. Your project context becomes machine-readable, persistent, and interoperable. This is foundational infrastructure for AI-context operations.
+
+**Key Features**
+
+Zero configuration required - works out of the box after installation. Includes 33+ tools for format operations. Operations average under 11 milliseconds. Synchronizes .faf files with CLAUDE.md automatically (keeping human-readable docs in sync with machine-readable persistent context). Built with 100% TypeScript strict mode. All 35 tests passing with production readiness confirmed.
 
 ---
 
-## 🚀 Quick Start
+## Scoring System
+
+Track your project's AI-readiness with a tiered scoring system:
 
+Trophy (100%) - Podium. Perfect AI and human balance. Gold (99%) - Gold standard. Silver (95-98%) - Excellence. Bronze (85-94%) - Production ready. Green (70-84%) - Good foundation. Yellow (55-69%) - Getting there. Red (0-54%) - Needs attention.
+
+Live output in Claude Desktop shows your score with a progress bar, current tier, and next milestone guidance.
+
+---
+
+## Quick Start
+
+Install globally via npm:
 ```bash
-# Install globally
 npm install -g claude-faf-mcp
+```
+
+Or via Homebrew:
+```bash
+brew install Wolfe-Jam/tap/claude-faf-mcp
+```
 
-# Add to Claude Desktop config
-# macOS/Linux: ~/Library/Application Support/Claude/claude_desktop_config.json
-# Windows: %APPDATA%\Claude\claude_desktop_config.json
+Add to Claude Desktop configuration. On macOS and Linux, edit ~/Library/Application Support/Claude/claude_desktop_config.json. On Windows, edit %APPDATA%\Claude\claude_desktop_config.json.
 
+```json
 {
   "mcpServers": {
     "claude-faf-mcp": {
@@ -80,207 +101,111 @@ npm install -g claude-faf-mcp
 }
 ```
 
-Restart Claude Desktop → Ready! 🏁
+Restart Claude Desktop to load the server.
 
 ---
 
-## 🏆 Anthropic-Approved MCP Server
-
-**claude-faf-mcp** is officially published in the [Anthropic MCP Registry](https://github.com/modelcontextprotocol/servers).
+## Scoring System Experience
 
-☑️ **Published to official Anthropic MCP registry** - Validated by Anthropic engineering team
-☑️ **[PR #2759](https://github.com/modelcontextprotocol/servers/pull/2759)** - Approved and available in registry
-☑️ **3,600+ total downloads** - Growing adoption across the Claude community
+This is what persistent project context looks like in action. When you run `faf_auto`, Claude scores your project's AI-readiness with a visual breakdown showing exactly where you stand and what to improve next.
 
-Registry listing: *"MCP server for .faf format. Context scoring engine with project context management."*
+![FAF Scoring Dashboard](https://cdn.jsdelivr.net/npm/claude-faf-mcp@latest/assets/faf-championship-scorecard.png)
 
-**The first .faf format server in the official Anthropic ecosystem.** 🧡⚡🏎️
+Live in Claude Desktop. Persistent across sessions. Your foundational context layer, measured and actionable.
 
 ---
 
-## 🎯 What is claude-faf-mcp?
-
-**The first and only AI Context MCP introducing Persistent Project Context.**
+## Available Tools
 
-MCP server that brings `.faf` format to Claude Desktop for instant project understanding.
+**Core Tools**
 
-**.faf = Foundational AI-Context Format | Project DNA ✨ for AI-Context, On-Demand**
+faf_init - Initialize project context. faf_auto - Auto-detect and populate context. faf_score - Calculate AI readiness. faf_status - Project health check.
 
-**Format-Driven Architecture:**
-- **Persistent** - Context survives across sessions, tools, and AIs
-- **Universal** - Works with any MCP client, CLI, workflow automation (n8n, Make, etc.), or AI assistant
-- **Stack-Agnostic** - Any language, framework, or setup
-- **Built with Claude** - Optimized for Claude Desktop while maintaining compatibility with ANY AI model or platform
+**Enhancement Tools**
 
-No limits. No restrictions. No boundaries. The only true format-first architecture for the AI era.
+faf_enhance - Optimize scoring. faf_sync - Sync files. faf_bi_sync - Bidirectional synchronization.
 
-### Key Features
+**File Operations**
 
-- ✅ **Zero Config** - Works out of the box
-- ✅ **33+ Tools** - Complete project management
-- ✅ **<11ms Speed** - Championship performance
-- ✅ **Bi-Sync** - .faf ↔ CLAUDE.md synchronization
-- ✅ **Type Safe** - 100% TypeScript strict mode
-- ✅ **Production Ready** - 35/35 tests passing
-
----
+faf_read - Read files. faf_write - Write files. faf_list - List directories. faf_search - Search file content.
 
-## 🛠️ Available Tools
+**Skills Integration**
 
-### Core Tools
-- `faf_init` - Initialize project context
-- `faf_auto` - Auto-detect and populate
-- `faf_score` - Calculate AI readiness
-- `faf_status` - Project health check
+faf_skills - List Claude Code skills from .faf file.
 
-### Enhancement Tools
-- `faf_enhance` - Optimize scoring
-- `faf_sync` - Sync files
-- `faf_bi_sync` - Bidirectional sync
-
-### File Operations
-- `faf_read` - Read files
-- `faf_write` - Write files
-- `faf_list` - List directories
-- `faf_search` - Search content
-
-### Skills & Integration
-- `faf_skills` - 🎸 List Claude Code skills from .faf file
-
-[See all 34+ tools →](https://faf.one/docs/tools)
+Full tool documentation available at https://faf.one/docs/tools.
 
 ---
 
-## 💡 Usage Example
+## Usage Example
 
-1. **Drop any project file** into Claude Desktop
-2. **Type**: "Run faf_auto to analyze this project"
-3. **Get instant context** - Claude understands your codebase
-4. **Use tools** - Access 33+ commands naturally in conversation
+1. Drop any project file into Claude Desktop
+2. Type: "Run faf_auto to analyze this project"
+3. Get instant context - Claude understands your codebase
+4. Access 33+ commands naturally in conversation
 
-The `.faf` file persists across conversations - no need to re-explain your project!
+The .faf file persists across conversations - no need to re-explain your project each time.
 
 ---
 
-## 📊 Technical Specs
-
-```
-Performance:   <11ms operations
-TypeScript:    100% strict mode
-Dependencies:  1 (MCP SDK only)
-Testing:       730 C.O.R.E empirical tests (part of 12,500+ FAF ecosystem validation)
-Build:         Zero errors
-Coverage:      4,400+ lines
-```
-
----
-
-## ✨ What's New
-
-### v2.6.2 - Documentation Improvements
-- **DRY Principle** - Version management simplified
-- **Single Source of Truth** - Version appears once in title
-- **Easier Maintenance** - Streamlined documentation updates
-
-### v2.6.1 - Official MCP Registry Publication
-- ☑️ **Published to Anthropic MCP Registry** - [PR #2759](https://github.com/modelcontextprotocol/servers/pull/2759) MERGED
-- **server.json** added for official registry listing
-- **mcpName** updated with correct capitalization format
-- **Registry validation** - First .faf format server in official ecosystem
-
-### v2.6.0 - Post-Evaluation Release (94.4/100 Gold Standard)
+## Technical Specifications
 
-🏆 **Post-Evaluation Release (94.4/100 Gold Standard)**
-- Type-safe tool handlers with proper TypeScript definitions
-- Community contribution framework (templates, guidelines, funding)
-- Repository cleanup: Removed 17K+ lines of legacy docs
-- Improved TypeScript strict mode compliance across all handlers
-
-### v2.5.2 - Visual Championship Experience
-- **Orange Smiley branding** - Complete visual identity on NPM
-- **Score card screenshot** - See the actual terminal output
-- **Championship polish** - Professional presentation
-
-### v2.5.1 - Documentation Polish
-- **Championship README** - Trophy section leads for immediate impact
-- **Optimized description** - Cleaner NPM presence
-- **Professional structure** - Scannable, modern layout
-
-### v2.5.0 - Championship Edition
-- **7-tier medal system** for AI-readiness scoring
-- **Visual progress bars** in terminal output
-- **Milestone tracking** with next-level guidance
-- **Enhanced scoring** for better project analysis
+Performance: Sub-11ms average operation time. TypeScript: 100% strict mode. Dependencies: 1 (MCP SDK only). Testing: 730 C.O.R.E empirical tests (part of 12,500+ FAF ecosystem validation). Build: Zero errors. Coverage: 4,400+ lines of code.
 
 ---
 
-## 🏗️ Development
+## Development
 
+Clone the repository:
 ```bash
-# Clone
 git clone https://github.com/Wolfe-Jam/claude-faf-mcp.git
 cd claude-faf-mcp
+```
 
-# Install & Build
+Install dependencies and build:
+```bash
 npm install
 npm run build
+```
 
-# Test
+Run tests:
+```bash
 npm test
+```
 
-# Link locally
+Link locally:
+```bash
 npm link
 ```
 
 ---
 
-## 📋 Requirements
+## Requirements
 
-- **Node.js** 18+
-- **Claude Desktop** (latest version)
-- **OS**: macOS, Linux, or Windows
+Node.js 18 or later. Claude Desktop (latest version). Operating system: macOS, Linux, or Windows.
 
 ---
 
-## 🔗 The FAF Ecosystem
+## The FAF Ecosystem
 
-- [🩵 **faf-cli**](https://npmjs.com/package/faf-cli) - Command line tool (v3.0.2 - Championship Edition 🏆 with Turbo Cat 😽)
-- [🧡 **claude-faf-mcp**](https://npmjs.com/package/claude-faf-mcp) - This MCP server
-- [💚 **faf.one**](https://faf.one) - Documentation & guides
-- [🖥️ **Chrome Extension**](https://chromewebstore.google.com/detail/lnecebepmpjpilldfmndnaofbfjkjlkm) - Browser integration
+faf-cli (npm) - Command line tool for local context management. claude-faf-mcp - This MCP server for Claude Desktop integration. faf.one - Documentation and guides. Chrome Extension - Browser integration for context collection.
 
 ---
 
-## 👤 Author
+## Author
 
-**James Wolfe (Wolfe-Jam)**
-Creator, .faf Format
-ORCID: [0009-0007-0801-3841](https://orcid.org/0009-0007-0801-3841)
+James Wolfe (Wolfe-Jam), creator of the .faf format. ORCID: 0009-0007-0801-3841.
 
 ---
 
-## 📄 License
+## License
 
-MIT License - See [LICENSE](./LICENSE) file
+MIT License. See LICENSE file for details.
 
-**Note**: The .faf-Engine is proprietary and available under separate license.
+Note: The .faf-Engine is proprietary and available under separate license.
 
 ---
 
-## 🤝 Contributing
-
-We welcome contributions! Join our [community discussions](https://github.com/Wolfe-Jam/claude-faf-mcp/discussions) or submit issues/PRs.
-
----
-
-<div align="center">
-
-**AI? Context? faf innit ✨**
-
-Made with 🧡 by wolfejam.dev
-
-**100% FREE Forever** • **Zero Dependencies** • **Zero Faff™**
-
-[⭐ Star on GitHub](https://github.com/Wolfe-Jam/claude-faf-mcp) • [📦 View on NPM](https://www.npmjs.com/package/claude-faf-mcp)
+## Contributing
 
-</div>
+Contributions are welcome. Join community discussions at https://github.com/Wolfe-Jam/claude-faf-mcp/discussions or submit issues and pull requests on GitHub.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-faf-mcp",
-  "version": "2.6.4",
+  "version": "2.6.6",
   "mcpName": "io.github.Wolfe-Jam/claude-faf-mcp",
   "description": "Persistent project context • MCP server for Claude Desktop • MIT License",
   "icon": "./assets/icons/faf-icon-64.png",
@@ -112,7 +112,7 @@
     "access": "public"
   },
   "dependencies": {
-    "@modelcontextprotocol/sdk": "^1.0.0"
+    "@modelcontextprotocol/sdk": "^1.20.1"
   },
   "devDependencies": {
     "@jest/globals": "^30.1.2",