Recursive Companion MCP

An MCP (Model Context Protocol) server that implements iterative refinement through self-critique cycles. Inspired by Hank Besser's recursive-companion, this implementation adds incremental processing to avoid timeouts and enable progress visibility.

Features

• Incremental Refinement: Avoids timeouts by breaking refinement into discrete steps

• Mathematical Convergence: Uses cosine similarity to measure when refinement is complete

• Domain-Specific Optimization: Auto-detects and optimizes for technical, marketing, strategy, legal, and financial domains

• Progress Visibility: Each step returns immediately, allowing UI updates

• Parallel Sessions: Support for multiple concurrent refinement sessions

• Auto Session Tracking: No manual session ID management needed

• AI-Friendly Error Handling: Actionable diagnostics and recovery hints for AI assistants

Related MCP server: Sequential Thinking Multi-Agent System

How It Works

The refinement process follows a Draft → Critique → Revise → Converge pattern:

1. Draft: Generate initial response

2. Critique: Create multiple parallel critiques (using faster models)

3. Revise: Synthesize critiques into improved version

4. Converge: Measure similarity and repeat until threshold reached

For detailed architecture diagrams and system design documentation, see ARCHITECTURE.md.

Installation

Prerequisites

• Python 3.10+

• uv package manager

• AWS Account with Bedrock access

• Claude Desktop app

Setup

1. Clone the repository:

2. Install dependencies:

3. Configure AWS credentials as environment variables or through AWS CLI

4. Add to Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

Basic Configuration:

Optimized Configuration (Recommended):

Performance Tips:

• Use Haiku for CRITIQUE_MODEL_ID for 50% speed improvement

• Lower CONVERGENCE_THRESHOLD to 0.95 for faster convergence

• Reduce PARALLEL_CRITIQUES to 2 for better resource usage

• See API_EXAMPLES.md for more configuration examples

Usage

The tool provides several MCP endpoints for iterative refinement:

Quick Start Examples

Simple refinement (auto-complete):

Step-by-step refinement (full control):

Session management:

Complete API Reference

For comprehensive examples with realistic scenarios, error handling patterns, and integration workflows, see API_EXAMPLES.md.

Available Tools

• start_refinement - Begin new refinement session with domain detection

• continue_refinement - Advance session through draft→critique→revise cycles

• get_final_result - Retrieve completed refinement

• get_refinement_status - Check progress without advancing

• current_session - Get active session info (no ID needed)

• list_refinement_sessions - See all active sessions

• abort_refinement - Stop refinement, return best version so far

• quick_refine - Auto-complete simple refinements (under 60s)

Configuration

🌐 Streamable HTTP Transport

The server supports a stateless Streamable HTTP transport for enterprise deployments requiring horizontal scalability and web-based integration.

Enable Streamable HTTP Transport

Or in a single command:

Streamable HTTP Features

• ✅ Stateless Request Handling: New server instance per request for complete isolation

• ✅ Session Persistence: Sessions maintained across requests via session IDs

• ✅ JSON-RPC 2.0 Compliant: Full MCP protocol support over HTTP

• ✅ Enterprise Scalability: Horizontal scaling with load balancers

• ✅ Web Integration: Perfect for web-based AI assistants

• ✅ Health Checks: Built-in health check endpoints

Example HTTP Request

Transport Modes Comparison

Performance

With optimized settings:

• Each iteration: 60-90 seconds

• Typical convergence: 2-3 iterations

• Total time: 2-4 minutes (distributed across multiple calls)

Using Haiku for critiques reduces iteration time by ~50%.

AI-Friendly Features

This tool includes special features for AI assistants using it:

• Auto Session Tracking: The current_session_id is automatically maintained

• Smart Error Messages: Errors include _ai_ prefixed fields with actionable diagnostics

• Performance Hints: Responses include optimization tips and predictions

• Progress Predictions: Convergence tracking includes estimates of remaining iterations

Example AI-helpful error response:

Architecture

Development

Running tests

Local testing

Automation Infrastructure

This project includes comprehensive automation for OSS readiness:

• 🤖 Dependabot: Automated dependency updates with intelligent grouping

• 🚀 Semantic Release: Automated versioning and releases based on conventional commits

• 🔒 Security Monitoring: Multi-tool security scanning (Safety, Bandit, CodeQL, Trivy)

• ✅ Quality Gates: Automated testing, coverage, linting, and type checking

• 📦 Dependency Management: Advanced dependency health monitoring and updates

Automation Commands

Development Workflow

1. Feature Development: Work on feature branches

2. Pull Requests: Quality gates run automatically

3. Code Review: Automated security and quality feedback

4. Merge to develop: Beta releases created automatically

5. Merge to main: Production releases created automatically

See AUTOMATION.md for complete automation documentation.

Attribution

This project is inspired by recursive-companion by Hank Besser. The original implementation provided the conceptual Draft → Critique → Revise → Converge pattern. This MCP version adds:

• Session-based incremental processing to avoid timeouts

• AWS Bedrock integration for Claude and Titan embeddings

• Domain auto-detection and specialized prompts

• Mathematical convergence measurement

• Support for different models for critiques vs generation

Contributing

Contributions are welcome! Please read our Contributing Guide for details.

License

MIT License - see LICENSE file for details.

Acknowledgments

• Original concept: Hank Besser's recursive-companion

• Built for the Model Context Protocol

• Uses AWS Bedrock for LLM access

• Inspired by iterative refinement patterns in AI reasoning