@brutalist/mcp 0.1.3 → 0.5.0

package/README.md

@@ -1,83 +1,261 @@
-# Brutalist MCP
+# Brutalist MCP 💀
 
-Your architecture will fail. Your startup will burn money. Your code has three unpatched CVEs.
+Your startup will fail. Your architecture will collapse. Your code is a security nightmare.
 
-All AIs are sycophants. This one doesn't lie.
+But this time, you'll know *why* before users do.
 
-## brutalist workflows
+## Deploy AI Critics That Don't Lie
+
+Every AI tells you what you want to hear. This one tells you what you need to know.
+
+Three brutal CLI agents. Zero sugar-coating. Maximum carnage.
+
+Three brutal CLI agents that can analyze anything. Each agent brings different perspectives to demolish your work from every angle.
+
+Real file-system analysis. Actual brutal prompts. Intelligent pagination for enterprise codebases. No participation trophies.
+
+## Brutalist Workflows
+
+### 🔍 **Codebase Destruction**
+
+> Analyze actual files in your repository for security holes, performance disasters, and architectural nightmares.
 
 ```bash
-# Destroy your architecture before users do
-roast_architecture "This microservices design for our startup..."
+# Demolish your entire codebase
+roast_codebase "/path/to/your/project"
 
-# Demolish your code quality with specific models
-roast_code(code="authentication.py", models=["google/gemini-2.5-pro", "anthropic/claude-3.5-sonnet"])
+# Target specific modules for focused brutality
+roast_codebase "/src/auth"          # Authentication vulnerabilities
+roast_codebase "/src/api/handlers"  # API endpoint disasters
+roast_codebase "/components"        # React component chaos
+```
+
+---
 
-# Reality check your ideas with 325+ AI models
-roast_idea "We're building a marketplace for..."
+### 💡 **Idea Obliteration**
 
-# Discover available models (325+ and growing)
-model_roster()  # Shows all available models
-model_roster(search="gemini")  # Find specific models
+> Reality-check your startup dreams, product concepts, and technical decisions.
 
-# Multi-model adversarial debate
-roast_debate "Should we use TypeScript or Go for this API?"
+```bash
+# Startup idea destruction
+roast_idea "A social network for developers to share code snippets"
+
+# Technical decision analysis
+roast_idea "Migrating our monolith to microservices with Kubernetes"
+
+# Product feature validation
+roast_idea "Adding AI-powered code suggestions to our IDE"
 ```
 
-## setup
+---
+
+### 🏗️ **Architecture Annihilation**
+
+> Find every scaling bottleneck, cost explosion, and operational nightmare in your system design.
 
 ```bash
-# VS Code/Cline one-liner
-code --add-mcp '{"name":"brutalist","command":"npx","args":["-y","@brutalist/mcp"],"env":{"OPENROUTER_API_KEY":"YOUR_KEY"}}'
+# System architecture review
+roast_architecture "Microservices with event sourcing and CQRS"
+
+# Infrastructure design analysis
+roast_architecture """
+API Gateway → Load Balancer → 3 Node.js services → PostgreSQL
+Redis for caching, Docker containers on AWS ECS
+"""
+```
+
+---
+
+### 🔒 **Security Demolition**
 
-# Claude Code
-claude mcp add brutalist --scope user -e OPENROUTER_API_KEY=YOUR_KEY -- npx -y @brutalist/mcp
+> Expose authentication bypasses, injection vulnerabilities, and data leak opportunities.
 
-# Gemini CLI
-gemini mcp add brutalist -e OPENROUTER_API_KEY=YOUR_KEY -- npx -y @brutalist/mcp
+```bash
+# Authentication system analysis
+roast_security "JWT tokens with user roles in localStorage"
 
-# Cursor: Add to ~/.cursor/mcp.json or use Settings > MCP & Integrations
-# Windsurf: Add to ~/.codeium/windsurf/mcp_config.json or use Plugin Store
+# API security review
+roast_security "GraphQL API with dynamic queries and no rate limiting"
 ```
 
-**Get key**: https://openrouter.ai/keys  
-**Models**: 325+ models dynamically fetched from OpenRouter. Always current.
+---
+
+### 🤺 **Multi-Agent Warfare**
 
-## why
+> Deploy multiple CLI agents in adversarial combat for maximum destruction.
 
-Every LLM defaults to "great idea!" because conflict doesn't pay. This deploys 325+ models to fight over your assumptions.
+```bash
+# Technical decision debate
+roast_cli_debate "Should we use TypeScript or Go for this API?"
+
+# Architecture comparison battle
+roast_cli_debate "Microservices vs Monolith for our e-commerce platform"
+```
+
+---
 
-## model selection
+### 🛠️ **Meta Commands**
 
 ```bash
-# Specific models
-roast_code(code="...", models=["google/gemini-2.5-pro", "openai/gpt-4o"])
+# Check which CLI agents are available
+cli_agent_roster()
+```
+
+## How It Works
+
+This MCP server orchestrates brutal feedback from locally installed CLI agents:
+- **Claude Code CLI** - Anthropic's code assistant with brutal system prompts
+- **Codex CLI** - OpenAI's code-focused model for technical criticism  
+- **Gemini CLI** - Google's model for architectural and system analysis
+
+Each agent runs locally on your machine with custom brutal prompts to find real problems before production fails.
+
+**⏱️ Analysis Timeout:** 25 minutes default - thorough analysis takes time to find real issues. Complex codebases and architectural reviews need deep analysis to catch subtle problems that quick scans miss.
+
+## Setup
+
+### Prerequisites
+
+Install at least one CLI agent:
+- **Claude Code**: `npm install -g claude` (or via Claude desktop app)
+- **Codex**: Install from [OpenAI Codex](https://github.com/openai/codex-cli)
+- **Gemini**: `npm install -g @google/gemini-cli` or authenticate via `gemini auth`
+
+<details>
+<summary><strong>Claude Code</strong> — One-liner</summary>
 
-# Random from 325+ models
-roast_idea "..."  # Chaos mode
+```bash
+claude mcp add brutalist --scope user -- npx -y @brutalist/mcp
+```
+</details>
+
+<details>
+<summary><strong>VS Code / Cline</strong> — Manual config</summary>
+
+```bash
+code --add-mcp '{"name":"brutalist","command":"npx","args":["-y","@brutalist/mcp"]}'
 ```
+</details>
+
+<details>
+<summary><strong>Gemini CLI</strong> — One-liner</summary>
+
+```bash
+gemini mcp add brutalist -- npx -y @brutalist/mcp
+```
+</details>
+
+<details>
+<summary><strong>Cursor</strong> — Manual config</summary>
 
-## tools
+Add to `~/.cursor/mcp.json` or use **Settings → MCP & Integrations**
 
-- **`roast_idea`** — Why imagination fails to become reality
-- **`roast_code`** — Security holes, performance disasters, maintainability nightmares
-- **`roast_architecture`** — Scaling failures, cost explosions, operational complexity
-- **`roast_research`** — Methodological flaws, irreproducible results, statistical crimes
-- **`roast_data`** — Overfitting, bias, correlation fallacies
-- **`roast_security`** — Attack vectors, authentication bypasses, data leaks
-- **`roast_product`** — UX disasters, adoption barriers, user abandonment
-- **`roast_infrastructure`** — Single points of failure, hidden costs, 3AM outages
-- **`roast_debate`** — Multiple models argue until truth emerges
-- **`model_roster`** — Browse and search 325+ available critics
+```json
+{
+  "brutalist": {
+    "command": "npx",
+    "args": ["-y", "@brutalist/mcp"]
+  }
+}
+```
+</details>
+
+<details>
+<summary><strong>Windsurf</strong> — Manual config</summary>
+
+Add to `~/.codeium/windsurf/mcp_config.json` or use **Plugin Store**
+
+```json
+{
+  "brutalist": {
+    "command": "npx",
+    "args": ["-y", "@brutalist/mcp"]
+  }
+}
+```
+</details>
+
+## 📄 Pagination Support (v0.5.0+)
+
+Handle enterprise-scale analyses that exceed Claude Code's 25K token limit:
+
+```bash
+# Enable pagination for large codebases
+roast_codebase({targetPath: "/monorepo", limit: 20000})
 
-## why this works
+# Continue reading from where you left off
+roast_codebase({targetPath: "/monorepo", offset: 20000, limit: 20000})
+
+# Smart chunking preserves readability
+roast_codebase({targetPath: "/complex-system", cursor: "offset:25000"})
+```
+
+**Features:**
+- **Smart Boundary Detection** - Preserves paragraphs and sentences
+- **Token Estimation** - Real-time cost awareness (~4 chars = 1 token)
+- **Rich Metadata** - Progress indicators and continuation instructions
+- **Configurable Chunks** - 1K to 100K characters per response
+
+## Tools
+
+### Code & Architecture Analysis
+| Tool | What gets destroyed | CLI Agents Used |
+|------|-------------------|-----------------|
+| `roast_codebase` | Security holes, performance disasters, maintainability nightmares in actual files | All available |
+| `roast_file_structure` | Directory chaos, naming disasters, structural nightmares | All available |
+| `roast_dependencies` | Version conflicts, security vulns, dependency hell | All available |
+| `roast_git_history` | Commit disasters, branching chaos, collaboration failures | All available |
+| `roast_test_coverage` | Testing gaps, quality blind spots, coverage lies | All available |
+
+### Conceptual Analysis
+| Tool | What gets destroyed | CLI Agents Used |
+|------|-------------------|-----------------|
+| `roast_idea` | Why imagination fails to become reality | All available |
+| `roast_architecture` | Scaling failures, cost explosions, operational complexity | All available |
+| `roast_research` | Methodological flaws, irreproducible results, statistical crimes | All available |
+| `roast_security` | Attack vectors, authentication bypasses, data leaks | All available |
+| `roast_product` | UX disasters, adoption barriers, user abandonment | All available |
+| `roast_infrastructure` | Single points of failure, hidden costs, 3AM outages | All available |
+
+### Meta Tools
+| Tool | What it does |
+|------|--------------|
+| `roast_cli_debate` | Multiple CLI agents argue until truth emerges |
+| `cli_agent_roster` | Shows which CLI agents are available on your system |
+
+## CLI Agent Selection
+
+The system automatically detects and uses available CLI agents:
+
+```bash
+# Use specific CLI agent
+roast_codebase(targetPath="/src", preferredCLI="claude")
+
+# Let system choose based on analysis type
+roast_security "/auth/module"  # Prefers Codex for security
+
+# Force multi-agent analysis (default)
+roast_idea "..."  # All available agents analyze in parallel
+```
+
+### Smart Selection Rules
+
+Different CLI agents excel at different analysis types:
+- **Code review**: Claude > Codex > Gemini
+- **Architecture**: Gemini > Claude > Codex  
+- **Security**: Codex > Claude > Gemini
+- **Research**: Claude > Gemini > Codex
+
+## Why This Works
 
 **Problem:** AI optimizes for engagement, not truth.  
-**Solution:** Deploy multiple models with conflicting incentives.  
-**Result:** Brutal honesty before expensive failures.
+**Solution:** Deploy multiple local CLI agents with adversarial perspectives.  
+**Result:** Brutal honesty through systematic destruction before expensive failures.
+
+Your code will fail. Your startup will struggle. Better to learn this from brutal CLI agents than from production outages at 3AM.
 
-Your code will fail. Your startup will struggle. Better to learn this from AI critics than users.
+The only AI that prevents disasters instead of causing them.
 
 ---
 
-OpenRouter API → 325+ models → Parallel execution → Adversarial synthesis
+Local CLI agents → Brutal system prompts → Parallel execution → Adversarial synthesis → Production survival

package/dist/brutalist-server.d.ts

@@ -3,14 +3,20 @@ import { BrutalistServerConfig } from './types/brutalist.js';
 export declare class BrutalistServer {
     server: McpServer;
     config: BrutalistServerConfig;
-    private openrouter;
+    private cliOrchestrator;
+    private httpTransport?;
     constructor(config?: BrutalistServerConfig);
+    private handleStreamingEvent;
+    private handleProgressUpdate;
     start(): Promise<void>;
+    private startStdioServer;
+    private startHttpServer;
     private registerTools;
-    private executeRoast;
-    private executeDebate;
+    private executeCLIDebate;
     private synthesizeDebate;
+    private executeBrutalistAnalysis;
     private formatToolResponse;
+    private formatPaginatedResponse;
     private formatErrorResponse;
     private handleToolExecution;
 }

package/dist/brutalist-server.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"brutalist-server.d.ts","sourceRoot":"","sources":["../src/brutalist-server.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAC;AAKpE,OAAO,EACL,qBAAqB,EAItB,MAAM,sBAAsB,CAAC;AAG9B,qBAAa,eAAe;IACnB,MAAM,EAAE,SAAS,CAAC;IAClB,MAAM,EAAE,qBAAqB,CAAC;IACrC,OAAO,CAAC,UAAU,CAAmB;gBAEzB,MAAM,GAAE,qBAA0B;IA0BxC,KAAK;IAWX,OAAO,CAAC,aAAa;YAkTP,YAAY;YAiCZ,aAAa;IAkC3B,OAAO,CAAC,gBAAgB;IA0BxB,OAAO,CAAC,kBAAkB;IAS1B,OAAO,CAAC,mBAAmB;YAUb,mBAAmB;CAUlC"}
+{"version":3,"file":"brutalist-server.d.ts","sourceRoot":"","sources":["../src/brutalist-server.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAC;AAQpE,OAAO,EACL,qBAAqB,EAKtB,MAAM,sBAAsB,CAAC;AAa9B,qBAAa,eAAe;IACnB,MAAM,EAAE,SAAS,CAAC;IAClB,MAAM,EAAE,qBAAqB,CAAC;IACrC,OAAO,CAAC,eAAe,CAAuB;IAC9C,OAAO,CAAC,aAAa,CAAC,CAAgC;gBAE1C,MAAM,GAAE,qBAA0B;IAwB9C,OAAO,CAAC,oBAAoB,CAgB1B;IAEF,OAAO,CAAC,oBAAoB,CAmB1B;IAEI,KAAK;YAeG,gBAAgB;YAMhB,eAAe;IAqE7B,OAAO,CAAC,aAAa;YAqhBP,gBAAgB;IAkK9B,OAAO,CAAC,gBAAgB;YAwFV,wBAAwB;IAuFtC,OAAO,CAAC,kBAAkB;IAwD1B,OAAO,CAAC,uBAAuB;IAwE/B,OAAO,CAAC,mBAAmB;YA8Bb,mBAAmB;CAUlC"}