czero-cc · arahangua · Aug 16, 2025 · Aug 16, 2025 · Aug 16, 2025
diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md
@@ -38,10 +38,12 @@ Fixes #(issue number)
 ## 🧪 Testing
 
 <!-- How did you test your changes? -->
-- [ ] Ran `uv run pytest`
-- [ ] Tested intelligent scraper with: `scapo scrape run --sources [source] --limit 5`
-- [ ] Verified LLM processing worked correctly
-- [ ] Checked that only AI/ML content was processed
+- [ ] Tested service discovery: `uv run scapo scrape discover --update`
+- [ ] Tested targeted scraping: `uv run scapo scrape targeted --service "[service]" --limit 10`
+- [ ] Tested batch processing: `uv run scapo scrape batch --category [category] --limit 10 --batch-size 2`
+- [ ] Verified LLM extraction quality (checked generated files in `models/`)
+- [ ] Tested TUI explorer: `uv run scapo tui`
+- [ ] For OpenRouter users: Updated context cache with `uv run scapo update-context`
 - [ ] Other testing (please describe):
 
 ## 📸 Screenshots (if applicable)

diff --git a/QUICKSTART.md b/QUICKSTART.md
@@ -23,6 +23,9 @@ cp .env.example .env
 LLM_PROVIDER=openrouter
 OPENROUTER_API_KEY=sk-or-v1-your-key-here  # Get from openrouter.ai
 OPENROUTER_MODEL=your_model
+
+# IMPORTANT: Update model context cache for accurate batching
+scapo update-context  # Without this, defaults to 4096 tokens (poor performance!)
 ```
 
 #### Option B: Ollama (Local)
@@ -117,27 +120,27 @@ scapo models search "copilot"         # Search for specific models
 cat models/audio/eleven-labs/cost_optimization.md
 ```
 
-### 5. (Optional) Use with Claude Desktop
+### 5. (Optional) MCP Server - Query Your Extracted Tips
+
+**Important:** The MCP server only reads tips you've already extracted. Run scrapers first (Steps 3-4) to populate models/ folder!
 
-Add SCAPO as an MCP server to query your extracted tips (from models/ folder) directly in Claude:
+Add SCAPO as an MCP server to query your extracted tips directly in MCP-compatible clients:
 
 ```json
-// Add to claude_desktop_config.json
+// Add to config.json
 {
   "mcpServers": {
     "scapo": {
       "command": "npx",
-      "args": ["@scapo/mcp-server"],
+      "args": ["@arahangua/scapo-mcp-server"],
       "env": {
-        "SCAPO_MODELS_PATH": "path/to/scapo/models"
+        "SCAPO_MODELS_PATH": "/path/to/scapo/models"
       }
     }
   }
 }
 ```
 
-Then ask Claude: "Get best practices for Midjourney" - no Python needed!
-
 ## 📊 Understanding the Output
 
 SCAPO creates organized documentation:
@@ -151,6 +154,19 @@ models/
 │       └── parameters.json     # Recommended settings
 ```
 
+## ⚙️ Utility Commands
+
+```bash
+# Update OpenRouter model context cache (for accurate batching)
+scapo update-context        # Updates if >24h old
+scapo update-context -f     # Force update
+
+# View extracted tips
+scapo tui                   # Interactive TUI explorer
+scapo models list           # List all extracted models
+scapo models search "copilot"  # Search for specific models
+```
+
 ## ⚙️ The --limit flag
 
 ```bash

diff --git a/README.md b/README.md
@@ -103,6 +103,13 @@ cp .env.example .env
 Get your API key from [openrouter.ai](https://openrouter.ai/)
 * you can also use local LLMs (Ollama, LMstudio). Check [QUICKSTART.md](./QUICKSTART.md)
 
+#### Important: Update Model Context Cache (OpenRouter users)
+```bash
+# REQUIRED for optimal performance - fetches accurate token limits
+scapo update-context  # Creates cache for faster processing
+```
+Without this, SCAPO defaults to 4096 tokens (severely limiting batch efficiency)
+
 
 ### 3. Start Extracting Optimization Tips
 
@@ -274,28 +281,24 @@ MAX_POSTS_PER_SCRAPE=100               # Limit per source
 ```
 Hand-wavy breakdown: With 5 posts, extraction success ~20%. With 20+ posts, success jumps to ~80%.
 
-## 🤖 MCP Server for Claude Desktop
+## 🤖 MCP Server (Optional Reader)
 
-Query your extracted tips directly in Claude (reads from models/ folder - run scrapers first!):
+**Note:** The MCP server is a reader that queries your already-extracted tips. You must run SCAPO scrapers first to populate the models/ folder!
 
 ```json
-// Add to %APPDATA%\Claude\claude_desktop_config.json (Windows)
-// or ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
+// Add to your client's mcp config.json 
 {
   "mcpServers": {
     "scapo": {
       "command": "npx",
-      "args": ["@scapo/mcp-server"],
+      "args": ["@arahangua/scapo-mcp-server"],
       "env": {
         "SCAPO_MODELS_PATH": "C:\\path\\to\\scapo\\models"  // Your models folder
       }
     }
   }
 }
 ```
-
-Then ask Claude: "Get me best practices for GitHub Copilot" or "What models are good for coding?"
-
 See [mcp/README.md](mcp/README.md) for full setup and available commands.
 
 ## 🎨 Interactive TUI
@@ -378,7 +381,12 @@ Built as part of the CZero Engine project to improve AI application development.
 - [OpenRouter](https://openrouter.ai/) for accessible AI APIs
 - Coffee ☕ for making this possible
 - [Ollama](https://ollama.com/) and [LMstudio](https://lmstudio.ai/) for awesome local LLM experience
-- [Awesome Generative AI](https://github.com/steven2358/awesome-generative-ai) & [Awesome AI Tools](https://github.com/mahseema/awesome-ai-tools) for service discovery
+- Service discovery powered by awesome lists:
+  - [steven2358/awesome-generative-ai](https://github.com/steven2358/awesome-generative-ai)
+  - [mahseema/awesome-ai-tools](https://github.com/mahseema/awesome-ai-tools)
+  - [filipecalegario/awesome-generative-ai](https://github.com/filipecalegario/awesome-generative-ai)
+  - [aishwaryanr/awesome-generative-ai-guide](https://github.com/aishwaryanr/awesome-generative-ai-guide)
+  - [eudk/awesome-ai-tools](https://github.com/eudk/awesome-ai-tools)
 - All opensource contributors in this space
 
 ---

diff --git a/mcp/README.md b/mcp/README.md
@@ -1,62 +1,76 @@
 # SCAPO MCP Server
 
-A Model Context Protocol (MCP) server for querying AI/ML best practices from the SCAPO (Stay Calm and Prompt On) knowledge base. Features intelligent fuzzy matching for improved user experience when querying model information.
+A Model Context Protocol (MCP) server that makes your locally-extracted SCAPO knowledge base queryable. 
+
+⚠️ **This is a reader, not a scraper!** You must first use [SCAPO](https://github.com/czero-cc/scapo) to extract tips into your `models/` folder.
 
 ## Documentation
 
 For comprehensive usage instructions, examples, and technical details, please see the **[Usage Guide](usage-guide.md)**.
 
-## Installation
+## Prerequisites
+
+1. **Clone and set up SCAPO first**:
+   ```bash
+   git clone https://github.com/czero-cc/scapo.git
+   cd scapo
+   # Follow SCAPO setup to run scrapers and populate models/
+   ```
+
+2. **Required**:
+   - Node.js 18+ 
+   - npm or npx
+   - Populated `models/` directory (from running SCAPO scrapers)
+
+## How It Works
+
+**IMPORTANT**: This MCP server ONLY reads from your local `models/` folder. It does NOT scrape data itself!
+
+1. First, use SCAPO to scrape and extract tips into `models/`
+2. Then, this MCP server makes those tips queryable in your AI client
 
-You can use this MCP server directly with `npx` (no Python required):
+## Quick Start
 
 ```bash
-npx @scapo/mcp-server
+# Step 1: Set up SCAPO and extract tips
+git clone https://github.com/czero-cc/scapo.git
+cd scapo
+# Follow SCAPO README to configure and run scrapers
+scapo scrape targeted --service "GitHub Copilot" --limit 20
+
+# Step 2: Configure MCP to read your extracted tips
+# Add to your MCP client config with YOUR path to scapo/models/
 ```
 
-Or install it globally:
+## Installation
 
 ```bash
-npm install -g @scapo/mcp-server
+npx @arahangua/scapo-mcp-server
 ```
 
-## Usage with Claude Desktop
-
-Add this to your Claude Desktop configuration file:
+## Configuration for MCP Clients
 
-### Windows
-Edit `%APPDATA%\Claude\claude_desktop_config.json`:
+Add this to your MCP client's configuration:
 
 ```json
 {
   "mcpServers": {
     "scapo": {
       "command": "npx",
-      "args": ["@scapo/mcp-server"],
+      "args": ["@arahangua/scapo-mcp-server"],
       "env": {
-        "SCAPO_MODELS_PATH": "C:\\path\\to\\scapo\\models"
+        "SCAPO_MODELS_PATH": "/absolute/path/to/your/scapo/models"  // From your cloned SCAPO repo!
       }
     }
   }
 }
 ```
 
-### macOS
-Edit `~/Library/Application Support/Claude/claude_desktop_config.json`:
+**Note:** Set `SCAPO_MODELS_PATH` to the absolute path of your SCAPO models directory.
 
-```json
-{
-  "mcpServers": {
-    "scapo": {
-      "command": "npx",
-      "args": ["@scapo/mcp-server"],
-      "env": {
-        "SCAPO_MODELS_PATH": "/path/to/scapo/models"
-      }
-    }
-  }
-}
-```
+For Claude Desktop specifically:
+- Windows: Edit `%APPDATA%\Claude\claude_desktop_config.json`
+- macOS: Edit `~/Library/Application Support/Claude/claude_desktop_config.json`
 
 ## Available Tools
 
@@ -89,7 +103,7 @@ List all available models by category.
 
 ```
 Arguments:
-- category: Model category ("text", "image", "video", "audio", "multimodal", "all")
+- category: Model category ("text", "image", "video", "audio", "multimodal", "code", "all")
 ```
 
 Example in Claude:
@@ -114,12 +128,14 @@ Example in Claude:
 ## Features
 
 - **Intelligent Fuzzy Matching**: Handles typos, partial names, and variations automatically
-- **No Python Required**: Pure Node.js implementation using npx
+  - Typo tolerance: `heygen` → "HeyGen", `gemeni` → "Gemini"
+  - Partial matching: `qwen` → finds all Qwen variants
+  - Case insensitive: `LLAMA-3` → "llama-3"
 - **Fully Standalone**: Works without any API server running
 - **Direct File Access**: Reads from local model files
 - **Smart Search**: Advanced search with similarity scoring
 - **Smart Recommendations**: Suggests models based on use case
-- **Easy Integration**: Works seamlessly with Claude Desktop
+- **Easy Integration**: Works with any MCP-compatible client
 - **Helpful Suggestions**: Provides alternatives when exact matches aren't found
 
 ## Use Cases
@@ -150,24 +166,13 @@ models/
 └── ...
 ```
 
-## Advantages Over Python Version
-
-1. **No Python Setup**: Works with just Node.js (which Claude Desktop already has)
-2. **Simple npx Usage**: One command to run, no installation needed
-3. **Better IDE Integration**: Works seamlessly with Cursor and other IDEs
-4. **Faster Startup**: Node.js starts faster than Python
-5. **Native JSON Handling**: Better performance for JSON operations
-
 ## Contributing
 
-To publish updates to npm:
-
-```bash
-cd mcp
-npm version patch  # or minor/major
-npm publish --access public
-```
+To contribute improvements:
+1. Fork the [SCAPO repository](https://github.com/czero-cc/SCAPO)
+2. Make your changes in the `mcp/` directory
+3. Submit a pull request
 
 ## License
 
-Same as the parent SCAPO repository.
+Same as the parent [SCAPO](https://github.com/czero-cc/SCAPO) repository.
diff --git a/mcp/package-lock.json b/mcp/package-lock.json
diff --git a/mcp/package.json b/mcp/package.json
@@ -1,6 +1,6 @@
 {
-  "name": "@scapo/mcp-server",
-  "version": "1.0.0",
+  "name": "@arahangua/scapo-mcp-server",
+  "version": "1.0.4",
   "description": "Stay Calm and Prompt On (SCAPO) - MCP server for AI/ML best practices",
   "main": "index.js",
   "type": "module",
@@ -27,4 +27,4 @@
   "engines": {
     "node": ">=18.0.0"
   }
-}
+}
diff --git a/mcp/usage-guide.md b/mcp/usage-guide.md
@@ -13,15 +13,11 @@ The SCAPO MCP (Model Context Protocol) server provides intelligent access to AI/
 ### Installation
 
 ```bash
-# Clone the repository
-git clone https://github.com/your-org/scapo.git
-cd scapo/mcp
+# After installing SCAPO, the MCP server can be used directly
+npx @arahangua/scapo-mcp-server
 
-# Install dependencies
-npm install
-
-# Or run directly with npx (no installation needed)
-npx @scapo/mcp-server
+# Or install globally
+npm install -g @arahangua/scapo-mcp-server
 ```
 
 ### Basic Setup
@@ -43,8 +39,8 @@ npm start
 {
   "mcpServers": {
     "scapo": {
-      "command": "node",
-      "args": ["/path/to/scapo/mcp/index.js"],
+      "command": "npx",
+      "args": ["@arahangua/scapo-mcp-server"],
       "env": {
         "SCAPO_MODELS_PATH": "/path/to/scapo/models"
       }

diff --git a/models/code/coderabbit/metadata.json b/models/code/coderabbit/metadata.json
@@ -0,0 +1,13 @@
+{
+  "service": "CodeRabbit",
+  "category": "code",
+  "last_updated": "2025-08-16T18:29:24.327526",
+  "extraction_timestamp": "2025-08-16T18:10:37.657193",
+  "data_sources": [
+    "Reddit API",
+    "Community discussions"
+  ],
+  "posts_analyzed": 104,
+  "confidence": "medium",
+  "version": "1.0.0"
+}
diff --git a/models/code/coderabbit/prompting.md b/models/code/coderabbit/prompting.md
@@ -0,0 +1,13 @@
+# CodeRabbit Prompting Guide
+
+*Last updated: 2025-08-16*
+
+## Tips & Techniques
+
+- Read more here: https://coderabbit.ai/blog/coderabbit-openai-rate-limits
+- This blog discusses how CodeRabbit uses FluxNinja Aperture to manage OpenAI’s limits and ensure optimal operation even during peak load.
+
+## Sources
+
+- Reddit community discussions
+- User-reported experiences