Reloaderoo
A local MCP server for developers that mirrors your in-development MCP server, allowing seamless restarts and tool updates so you can build, test, and iterate on your MCP server within the same AI session without interruption.

Test and hot-reload MCP servers with CLI inspection tools and transparent proxy capabilities
A dual-mode MCP development tool that operates as both a CLI inspection tool and a transparent proxy server for the Model Context Protocol (MCP). Works excellently with VSCode MCP, well with Claude Code, and supports other MCP-enabled clients.
🔄 Two Modes, One Tool
reloaderoo provides two distinct operational modes to fit different development workflows:
🔍 CLI Mode (Inspection & Testing)
Direct command-line access to MCP servers without client setup:
- ✅ One-shot commands - Test tools, list resources, get server info
- ✅ No MCP client required - Perfect for testing and debugging
- ✅ Raw JSON output - Ideal for scripts and automation
- ✅ 8 inspection commands - Complete MCP protocol coverage
🔄 Proxy Mode (Hot-Reload Development)
Transparent proxy server that enables seamless hot-reloading:
- ✅ Hot-reload MCP servers without disconnecting your AI client
- ✅ Session persistence - Keep your development context intact
- ✅ Automatic
restart_servertool - AI agents can restart servers on demand - ✅ Transparent forwarding - Full MCP protocol passthrough
🎯 Why reloaderoo?
When developing MCP servers, you typically face two problems:
- Testing requires complex MCP client setup → CLI mode solves this
- Code changes require restarting your entire AI session → Proxy mode solves this
Both modes work together to create a seamless MCP development experience.
🚀 Quick Start
Installation
# Install globally for easy access
npm install -g reloaderoo
# Or use with npx (no installation required)
npx reloaderoo --help
Choose Your Mode
🔍 CLI Mode - Testing & Debugging
Perfect for testing MCP servers without client setup:
# List all tools in your server
reloaderoo inspect list-tools -- node my-mcp-server.js
# Call a specific tool
reloaderoo inspect call-tool echo --params '{"message":"hello"}' -- node my-mcp-server.js
# Get server information
reloaderoo inspect server-info -- node my-mcp-server.js
🔄 Proxy Mode - Hot-Reload Development
For AI client integration with hot-reload capabilities:
# Start proxy server (your AI client connects to this)
reloaderoo proxy -- node my-mcp-server.js
# With debug logging
reloaderoo proxy --log-level debug -- node my-mcp-server.js
Then configure your AI client to connect to reloaderoo instead of directly to your server.
🎯 Recommended Clients
Best Experience: VSCode & Cursor - Full protocol support with automatic capability detection Good Experience: Claude Code & Windsurf - Works well, may need manual refresh for new tools
🛠️ Development Workflow
🔍 CLI Mode Workflow (Testing & Debugging)
Perfect for testing individual tools or debugging server issues:
# 1. Test your server quickly
reloaderoo inspect list-tools -- node my-mcp-server.js
# 2. Call specific tools to verify behavior
reloaderoo inspect call-tool my_tool --params '{"param":"value"}' -- node my-mcp-server.js
# 3. Check server health
reloaderoo inspect ping -- node my-mcp-server.js
🔄 Proxy Mode Workflow (Hot-Reload Development)
For full development sessions with AI clients:
1. Start Development Session
Configure your AI client to connect to reloaderoo proxy instead of your server directly:
reloaderoo proxy -- node my-mcp-server.js
# or with debug logging:
reloaderoo proxy --log-level debug -- node my-mcp-server.js
2. Develop Your MCP Server
Work on your server code as usual:
// my-mcp-server.js
export const server = new Server({
name: "my-awesome-server",
version: "1.0.0"
});
// Add new tools, modify existing ones, etc.
server.addTool("new_feature", /* ... */);
3. Test Changes Instantly
Ask your AI agent to restart the server:
"Please restart the MCP server to load my changes"
The agent will call the restart_server tool automatically. Your new capabilities are immediately available!
4. Continue Development
Your AI session continues with the updated server capabilities. No connection loss, no context reset.
📋 Command Line Interface
reloaderoo provides two primary modes of operation:
reloaderoo [options] [command]
Global Options:
-V, --version Output the version number
-h, --help Display help for command
Commands:
proxy [options] -- <command> 🔄 Run as MCP proxy server (hot-reload mode)
inspect [subcommand] 🔍 Inspect and debug MCP servers (CLI mode)
info [options] 📊 Display version and configuration information
help [command] ❓ Display help for command
🔄 Proxy Mode Commands (Hot-Reload Development)
reloaderoo proxy [options] -- <child-command> [child-args...]
Options:
-w, --working-dir <directory> Working directory for the child process
-l, --log-level <level> Log level (debug, info, notice, warning, error, critical)
-f, --log-file <path> Custom log file path (logs to stderr by default)
-t, --restart-timeout <ms> Timeout for restart operations (default: 30000ms)
-m, --max-restarts <number> Maximum restart attempts 0-10 (default: 3)
-d, --restart-delay <ms> Delay between restart attempts (default: 1000ms)
-q, --quiet Suppress non-essential output
--no-auto-restart Disable automatic restart on crashes
--debug Enable debug mode with verbose logging
--dry-run Validate configuration without starting proxy
Examples:
reloaderoo proxy -- node server.js
reloaderoo -- node server.js # Same as above (proxy is default)
reloaderoo proxy --log-level debug -- python mcp_server.py --port 8080
🔍 CLI Mode Commands (Inspection & Testing)
reloaderoo inspect [subcommand] [options] -- <child-command> [child-args...]
Subcommands:
server-info [options] Get server information and capabilities
list-tools [options] List all available tools
call-tool [options] <name> Call a specific tool
list-resources [options] List all available resources
read-resource [options] <uri> Read a specific resource
list-prompts [options] List all available prompts
get-prompt [options] <name> Get a specific prompt
ping [options] Check server connectivity
mcp [options] Start MCP inspection server (exposes debug tools as MCP server)
Common Options (available for all subcommands):
-w, --working-dir <dir> Working directory for the child process
-t, --timeout <ms> Operation timeout in milliseconds (default: 30000)
-q, --quiet Suppress child process stderr output (get clean JSON)
Examples:
reloaderoo inspect list-tools -- node server.js
reloaderoo inspect call-tool get_weather --params '{"location": "London"}' -- node server.js
reloaderoo inspect server-info -- node server.js
reloaderoo inspect mcp -- node server.js # Start MCP inspection server
# Get clean JSON output without server logs
reloaderoo inspect list-tools --quiet -- node server.js
reloaderoo inspect call-tool echo --quiet --params '{"message":"test"}' -- node server.js
Info Command (Diagnostics)
reloaderoo info [options]
Options:
--verbose Show detailed system information
Examples:
reloaderoo info # Show basic system information
reloaderoo info --verbose # Show detailed diagnostics
🔍 CLI Mode Deep Dive (Inspection & Testing)
CLI mode provides direct command-line access to MCP servers without requiring client setup - perfect for testing and debugging.
🤖 AI Agent Use Case - The Primary Design Goal
CLI mode is specifically designed for AI agents (like Claude Code, Cursor, etc.) that have terminal access but don't have MCP server configuration capabilities. This solves a critical development workflow problem:
The Problem: When an AI agent is helping you develop an MCP server, it needs to test changes, but:
- ❌ The agent can't configure itself to use your MCP server directly
- ❌ Asking users to manually configure MCP clients breaks the development flow
- ❌ Using resource tools or web fetching is indirect and limited
The Solution: CLI mode gives AI agents direct, terminal-based access to your MCP server:
- ✅ No Client Configuration: Agent uses terminal commands, not MCP client setup
- ✅ Stateless & Reliable: Each command runs independently - no persistent connections to fail
- ✅ Raw Protocol Access: Agent sees exact MCP inputs/outputs for transparent debugging
- ✅ Immediate Testing: Agent can validate changes instantly without user intervention
🔧 Technical Benefits
Stateless Execution:
- Each CLI command spawns the server, executes the request, and terminates
- Perfect reliability - no persistent state to get corrupted
- No connection management or session handling complexity
⚠️ Important Limitation:
- Servers with in-memory state machines won't work properly in CLI mode
- Each command is isolated - no shared state between calls
- For stateful servers, use Proxy mode instead
Transparent Debugging:
- Raw JSON output shows exact MCP protocol requests/responses
- No proxy layer or client interpretation
- Perfect for understanding what's actually happening at the protocol level
- Use
--quietflag to suppress server logs and get clean JSON for scripting
📝 Direct CLI Commands (One-shot execution)
Execute single commands and get immediate results:
# List all tools in your server
reloaderoo inspect list-tools -- node my-server.js
# Call a specific tool
reloaderoo inspect call-tool echo --params '{"message":"hello"}' -- node my-server.js
# Get server information
reloaderoo inspect server-info -- node my-server.js
# Check server connectivity
reloaderoo inspect ping -- node my-server.js
# Get clean JSON output without server logs (perfect for scripting)
reloaderoo inspect list-tools --quiet -- node my-server.js
reloaderoo inspect call-tool echo --quiet --params '{"message":"hello"}' -- node my-server.js
🔧 MCP Inspection Server (Persistent CLI mode for MCP clients)
Start CLI mode as a persistent MCP server for interactive debugging:
# Start reloaderoo in CLI mode as an MCP server
reloaderoo inspect mcp -- node my-server.js
This runs CLI mode as a persistent MCP server, exposing 8 debug tools through the MCP protocol:
list_tools- List all server toolscall_tool- Call any server toollist_resources- List all server resourcesread_resource- Read any server resourcelist_prompts- List all server promptsget_prompt- Get any server promptget_server_info- Get comprehensive server infoping- Test server connectivity
🏗️ How Both Modes Work
🔄 Proxy Mode Architecture (Hot-Reload Development)
graph LR
A[AI Client] -->|MCP Protocol| B[reloaderoo proxy]
B -->|Forwards Messages| C[Your MCP Server]
B -->|Manages Lifecycle| C
B -->|Adds restart_server Tool| A
style A fill:#e1f5fe
style B fill:#f3e5f5
style C fill:#e8f5e8
Proxy Mode Magic:
- Transparent Forwarding - All MCP messages pass through seamlessly
- Capability Augmentation - Adds
restart_servertool to your server's capabilities - Process Management - Spawns, monitors, and restarts your server process
- Session Persistence - Client connection remains active during server restarts
- Protocol Compliance - Full MCP v2025-03-26 support with intelligent fallbacks
🔍 CLI Mode Architecture (Direct Testing)
graph LR
A[Your Terminal] -->|Direct Commands| B[reloaderoo inspect]
B -->|Spawns & Queries| C[Your MCP Server]
B -->|Returns JSON| A
style A fill:#e8f5e8
style B fill:#f3e5f5
style C fill:#e1f5fe
CLI Mode Magic:
- Direct Execution - No proxy layer, direct command execution
- One-Shot Queries - Each command spawns server, executes, and returns results
- Raw JSON Output - Perfect for automation and scripting
- No Client Setup - Test MCP servers without configuring MCP clients
- 8 Inspection Commands - Complete MCP protocol coverage for testing
🔧 Configuration
Environment Variables
Configure reloaderoo behavior via environment variables:
# Logging Configuration
export MCPDEV_PROXY_LOG_LEVEL=debug # Log level (debug, info, notice, warning, error, critical)
export MCPDEV_PROXY_LOG_FILE=/path/to/log # Custom log file path (default: stderr)
export MCPDEV_PROXY_DEBUG_MODE=true # Enable debug mode (true/false)
# Process Management
export MCPDEV_PROXY_RESTART_LIMIT=5 # Maximum restart attempts (0-10, default: 3)
export MCPDEV_PROXY_AUTO_RESTART=true # Enable/disable auto-restart (true/false)
export MCPDEV_PROXY_TIMEOUT=30000 # Operation timeout in milliseconds
export MCPDEV_PROXY_RESTART_DELAY=1000 # Delay between restart attempts in milliseconds
export MCPDEV_PROXY_CWD=/path/to/directory # Default working directory
🎨 Integration Examples
🔄 Proxy Mode Integration (MCP Client Setup)
Configure your MCP client to connect to reloaderoo proxy instead of your server directly:
{
"mcpServers": {
"my-dev-server": {
"command": "reloaderoo",
"args": [
"proxy",
"--",
"node",
"my-dev-server.js"
]
}
}
}
🔍 CLI Mode Integration (Automation & Testing)
Perfect for CI/CD, testing scripts, and automation workflows:
#!/bin/bash
# Example: Test script for your MCP server
# Check if server is healthy (use --quiet for clean output)
if reloaderoo inspect ping --quiet -- node my-server.js >/dev/null 2>&1; then
echo "✅ Server is healthy"
else
echo "❌ Server health check failed"
exit 1
fi
# Test specific functionality with clean JSON output
echo "Testing echo tool..."
result=$(reloaderoo inspect call-tool echo --quiet --params '{"message":"test"}' -- node my-server.js)
# Parse and validate JSON response (no server logs to interfere)
if echo "$result" | jq -e '.content[0].text' >/dev/null; then
echo "✅ Echo tool test passed"
echo "Response: $(echo "$result" | jq -r '.content[0].text')"
else
echo "❌ Echo tool test failed"
echo "Raw response: $result"
exit 1
fi
# List tools and count them
echo "Checking available tools..."
tools_count=$(reloaderoo inspect list-tools --quiet -- node my-server.js | jq '.tools | length')
echo "✅ Found $tools_count tools available"
🚨 Troubleshooting
🔄 Proxy Mode Issues
Server won't start in proxy mode:
# Check if your server runs independently first
node my-dev-server.js
# Then try with reloaderoo proxy to validate configuration
reloaderoo proxy -- node my-dev-server.js
Connection problems with MCP clients:
# Enable debug logging to see what's happening
reloaderoo proxy --log-level debug -- node my-server.js
# Check system info and configuration
reloaderoo info --verbose
Restart failures in proxy mode:
# Increase restart timeout
reloaderoo proxy --restart-timeout 60000 -- node my-server.js
# Check restart limits
reloaderoo proxy --max-restarts 5 -- node my-server.js
🔍 CLI Mode Issues
CLI commands failing:
# Test basic connectivity first
reloaderoo inspect ping -- node my-server.js
# Enable debug logging for CLI commands
reloaderoo inspect list-tools --log-level debug -- node my-server.js
JSON parsing errors:
# Use --raw flag to see unformatted output
reloaderoo inspect server-info --raw -- node my-server.js
# Ensure your server outputs valid JSON
node my-server.js | head -10
General Debug Mode
# Get detailed information about what's happening
reloaderoo proxy --debug -- node my-server.js # For proxy mode
reloaderoo inspect list-tools --log-level debug -- node my-server.js # For CLI mode
# View system diagnostics
reloaderoo info --verbose
🤝 Contributing
We welcome contributions! Please see CONTRIBUTING.md for guidelines.
Development Setup
git clone https://github.com/cameroncooke/reloaderoo.git
cd reloaderoo
npm install
npm run build
npm test
Running Tests
npm run test # All tests (unit, integration, E2E)
npm run test:unit # Unit tests only
npm run test:integration # Integration tests only
npm run test:e2e # End-to-end tests only
npm run test:coverage # Test coverage report
Testing Guidelines: See docs/TESTING_GUIDELINES.md for comprehensive testing standards and best practices.
📄 License
MIT License - see LICENSE file for details.
🔗 Related Projects
- XcodeBuildMCP - MCP server for Xcode development workflow automation
Verwandte Server
Alpha Vantage MCP Server
SponsorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
stdout-mcp-server
Captures and manages stdout logs from multiple processes via a named pipe system for real-time debugging and analysis.
Claude-FAF-MCP
Only Persistent Project Context MCP Server - Official Anthropic Registry
Gemini CLI
Integrates with the unofficial Google Gemini CLI, allowing file access within configured directories.
Remote MCP Server (Authless)
An example of a remote MCP server deployable on Cloudflare Workers without authentication, allowing for custom tool integration.
Facets Module
Create and manage Terraform modules for cloud-native infrastructure using the Facets.cloud FTF CLI.
Draw Architecture
Generate draw.io system architecture diagrams from text descriptions using the ZhipuAI large model.
Remote MCP Server (Authless)
A remote MCP server deployable on Cloudflare Workers without authentication.
ComfyUI
An MCP server for ComfyUI integration.
VULK MCP Server
Build, edit, and deploy full-stack web applications from any AI assistant. 9 MCP tools with real AI generation via SSE streaming.
BCMS MCP
Give me a one - two sentence description of the BCMS MCP # MCP The BCMS Model Context Protocol (MCP) integration enables AI assistants like Claude, Cursor, and other MCP-compatible tools to interact directly with your BCMS content. This allows you to create, read, and update content entries, manage media files, and explore your content structure—all through natural language conversations with AI. ## What is MCP? The [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) is an open standard developed by Anthropic that allows AI applications to securely connect to external data sources and tools. With BCMS MCP support, you can leverage AI assistants to: - Query and explore your content structure - Create new content entries with AI-generated content - Update existing entries - Manage your media library - Get intelligent suggestions based on your content model --- ## Getting Started ### Prerequisites 1. A BCMS account with an active instance 2. An MCP key with appropriate permissions 3. An MCP-compatible client (Claude Desktop, Cursor, or any MCP client) ### Step 1: Create an MCP Key 1. Navigate to your BCMS dashboard 2. Go to Settings → MCP 3. Click Create MCP Key 4. Configure the permissions for templates you want the AI to access:GET: Read entries 5. POST: Create entries 6. PUT: Update entries 7. DELETE: Delete entries Note: Right now, MCP only supports creating, reading and updating content. ### Step 2: Configure Your MCP Client You can find full instructions for integrating BCMS with your AI tools right inside BCMS, on the MCP page. But in general, installing BCMS MCP works in a standard way: ``` { "mcpServers": { "bcms": { "url": "https://app.thebcms.com/api/v3/mcp?mcpKey=YOUR_MCP_KEY" } } } ``` ## Available Tools Once connected, your AI assistant will have access to the following tools based on your MCP key permissions: ### Content Discovery #### list_templates_and_entries Lists all templates and their entries that you have access to. This is typically the first tool to call when exploring your BCMS content. Returns: - Template IDs, names, and slugs - Entry IDs with titles and slugs for each language Example prompt: "Show me all the templates and entries in my BCMS" --- ### Entry Management #### list_entries_for_{templateId} Retrieves all entries for a specific template with full content data. A separate tool is generated for each template you have access to. Returns: - Complete entry data including all meta fields - Content in all configured languages - Entry statuses Example prompt: "List all blog posts from my Blog template" --- #### create_entry_for_{templateId} Creates a new entry for a specific template. The input schema is dynamically generated based on your template's field structure. Input: - statuses: Array of status assignments per language - meta: Array of metadata for each language (title, slug, custom fields) - content: Array of content nodes for each language Example prompt: "Create a new blog post titled 'Getting Started with BCMS' with a brief introduction paragraph" --- #### update_entry_for_{templateId} Updates an existing entry for a specific language. Input: - entryId: The ID of the entry to update - lng: Language code (e.g., "en") - status: Optional status ID - meta: Updated metadata - content: Updated content nodes Example prompt: "Update the introduction paragraph of my 'Getting Started' blog post" --- ### Media Management #### list_all_media Lists all media files in your media library. Returns: - Media IDs, names, and types - File metadata (size, dimensions for images) - Parent directory information Example prompt: "Show me all images in my media library" --- #### list_media_dirs Lists the directory structure of your media library. Returns: - Hierarchical directory structure - Directory IDs and names Example prompt: "Show me the folder structure of my media library" --- #### create-media-directory Creates a new directory in your media library. Input: - name: Name of the directory - parentId: Optional parent directory ID (root if not specified) Example prompt: "Create a new folder called 'Blog Images' in my media library" --- #### request-upload-media-url Returns a URL you use to upload a file (for example via POST with multipart form data), which avoids pushing large binaries through the MCP tool payload. You still need a valid file name and MIME type when uploading, as described in the tool response. Availability: Only when the MCP key has Can mutate media enabled. Example prompt: “Give me an upload URL for a new hero image, then tell me how to upload it.” Input: - fileName: Name of the file with extension - fileData: Base64-encoded file data (with data URI prefix) - parentId: Optional parent directory ID Example prompt: "Upload this image to my Blog Images folder" --- ### Linking Tools #### get_entry_pointer_link Generates an internal BCMS link to an entry for use in content. Input: - entryId: The ID of the entry to link to Returns: - Internal link format: entry:{entryId}@*_{templateId}:entry Example prompt: "Get me the internal link for the 'About Us' page entry" --- #### get_media_pointer_link Generates an internal BCMS link to a media item for use in content. Input: - mediaId: The ID of the media item Returns: - Internal link format: media:{mediaId}@*_@*_:entry Example prompt: "Get the link for the hero image so I can use it in my blog post" --- ## Content Structure ### Entry Content Nodes When creating or updating entries, content is structured as an array of nodes. Supported node types include: Type Description paragraph Standard text paragraph heading Heading (h1-h6) bulletList Unordered list orderedList Numbered list listItem List item codeBlock Code block with syntax highlighting blockquote Quote block image Image node widget Custom widget with props ### Example Content Structure ``` { "content": [ { "lng": "en", "nodes": [ { "type": "heading", "attrs": { "level": 1 }, "content": [ { "type": "text", "text": "Welcome to BCMS" } ] }, { "type": "paragraph", "content": [ { "type": "text", "text": "This is your first paragraph." } ] } ] } ] } ``` ## Security & Permissions ### MCP Key Scopes Your MCP key controls what the AI can access: - Template Access: Only templates explicitly granted in the MCP key are visible - Operation Permissions: Each template can have independent GET/POST/PUT/DELETE permissions - Media Access: Media operations are controlled separately ### Best Practices 1. Principle of Least Privilege: Only grant the permissions needed for your use case 2. Separate Keys: Create different MCP keys for different purposes or team members 3. Regular Rotation: Periodically rotate your MCP keys ## Use Cases ### Content Creation Workflows Blog Post Creation "Create a new blog post about the benefits of headless CMS. Include an introduction, three main benefits with explanations, and a conclusion. Use the Blog template." Product Updates "Update the price field for all products in the Electronics category to apply a 10% discount" ### Content Exploration Content Audit "List all blog posts that don't have a featured image set" Translation Status "Show me which entries are missing German translations" ### Media Organization Library Cleanup "Show me all unused images in the media library" Folder Setup "Create folder structure for: Products > Categories > Electronics, Clothing, Home" ## Troubleshooting ### Common Issues #### "MCP key not found" - Verify your MCP key format: keyId.keySecret.instanceId - Ensure the MCP key hasn't been deleted or deactivated - Check that you're using the correct instance #### "MCP key does not have access to template" - Review your MCP key permissions in the dashboard - Ensure the required operation (GET/POST/PUT/DELETE) is enabled for the template #### Session Expired - MCP sessions may timeout after periods of inactivity - Simply start a new conversation to establish a fresh session ### Getting Help - Documentation: [thebcms.com/docs](https://thebcms.com/docs) - Support: [[email protected]](mailto:[email protected]) - Community: [Join BCMS Discord](https://discord.com/invite/SYBY89ccaR) for community support ## Technical Reference ### Endpoint POST https://app.thebcms.com/api/v3/mcp?mcpKey={MCP_KEY} ### Transport BCMS MCP uses the Streamable HTTP transport with session management. Sessions are maintained via the mcp-session-id header. ### Response Format All tools return structured JSON responses conforming to the MCP specification with: - content: Array of content blocks - structuredContent: Typed response data ## Rate Limits MCP requests are subject to the same rate limits as API requests: - Requests are tracked per MCP key - Contact support if you need higher limits for production workloads