1. Conduid
  2. AI
  3. MCP
MCP server · AI

MCP

A crate for making MCP (Model Context Protocol) compatible programs with rust

Unclaimed last commit a year ago model-context-protocolllmai
52Fair

Scored 5 months ago · breakdown

About MCP

MCP is an MCP server published by EmilLindfors in the AI category: a crate for making MCP (Model Context Protocol) compatible programs with rust.

The repository has 18 stars and 10 forks, with the last commit a year ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/EmilLindfors/mcp

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about MCP

Powered by Claude · Grounded in docs

I know everything about MCP. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

Static checks derived from the repository 6 hours ago. Heuristics on the source, not an audit.

  • ✓README presentThe repository ships documentation.
  • !License declaredNo license file. Legally you have no right to use it in a product.
  • ✓Tests presentA test suite exists in the repository.
  • !Dependencies pinnedNo lockfile. Installs can pull different code over time.
  • ✓No dynamic code executionSource scan found no eval, dynamic Function, or shell execution.
  • !Scoped permissionsDoesn't declare a permission scope. Assume it can do anything its process can.

README

mcp.rs

A Rust implementation of the Model Context Protocol (MCP), providing a standardized way for AI models to access external context and resources.

Quickstart

Client

# List resources
cargo run --bin client list-resources

# Read a specific file
cargo run --bin client read-resource -u "file:///path/to/file"

# Use a prompt
cargo run --bin client get-prompt -n "code_review" -a '{"code": "fn main() {}", "language": "rust"}'

# Call a tool
cargo run --bin client -- --server "http://127.0.0.1:3000" call-tool --name "file_system" --args '{\"operation\": \"read_file\", \"path\": \"Config.toml\"}'

# Set log level
cargo run --bin client set-log-level -l "debug"

# Use SSE transport
cargo run --bin client -t sse -s http://localhost:3000 list-resources

Server

# Run with test config
cargo run --bin server -- --config "../servers/test.json"

Overview

mcp.rs is a high-performance, type-safe Rust implementation of the Model Context Protocol, designed to enable seamless communication between AI applications and their integrations. It provides a robust foundation for building MCP servers that can expose various types of resources (files, data, APIs) to AI models.

Features

  • Multiple Transport Types:

    • Standard Input/Output (stdio) transport for CLI tools
    • HTTP with Server-Sent Events (SSE) for web integrations
    • Extensible transport system for custom implementations
  • Resource Management:

    • File system resource provider
    • Resource templating support
    • Real-time resource updates
    • Resource subscription capabilities
  • Flexible Configuration:

    • YAML/JSON configuration files
    • Environment variable overrides
    • Command-line arguments
    • Sensible defaults
  • Security:

    • Built-in access controls
    • Path traversal protection
    • Rate limiting
    • CORS support

Installation

Add mcp.rs to your project's Cargo.toml:

[dependencies]
mcp = "0.1.0"

Quick Start

  1. Create a basic MCP server:
use mcp::{McpServer, ServerConfig};

#[tokio::main]
async fn main() -> Result<(), mcp::error::McpError> {
    // Create server with default configuration
    let server = McpServer::new(ServerConfig::default());
    
    // Run the server
    server.run().await
}
  1. Configure via command line:
# Run with stdio transport
mcp-server -t stdio

# Run with SSE transport on port 3000
mcp-server -t sse -p 3000

# Enable debug logging
mcp-server -l debug
  1. Or use a configuration file:
server:
  name: "my-mcp-server"
  version: "1.0.0"
  transport: sse
  port: 3000

resources:
  root_path: "./resources"
  allowed_schemes:
    - file
  max_file_size: 10485760

security:
  enable_auth: false
  allowed_origins:
    - "*"

logging:
  level: "info"
  format: "pretty"

Architecture

mcp.rs follows a modular architecture:

  • Transport Layer: Handles communication between clients and servers
  • Protocol Layer: Implements the MCP message format and routing
  • Resource Layer: Manages access to external resources
  • Configuration: Handles server settings and capabilities

Configuration Options

Option Description Default
transport Transport type (stdio, sse) stdio
port Server port for network transports 3000
log_level Logging level info
resource_root Root directory for resources ./resources

API Reference

For detailed API documentation, run:

cargo doc --open

Examples (TODO)

Check out the /examples directory for:

  • Basic server implementation
  • Custom resource provider
  • Configuration examples
  • Integration patterns

Contributing

Contributions are welcome! Please read our Contributing Guidelines before submitting pull requests.

Development Requirements

  • Rust 1.70 or higher
  • Cargo
  • Optional: Docker for containerized testing

Running Tests

# Run all tests
cargo test

# Run with logging
RUST_LOG=debug cargo test

License

This project is licensed under the MIT License.

Acknowledgments

This implementation is based on the Model Context Protocol specification and inspired by the reference implementation.

Contact

README fetched from the repository 6 hours ago. The original is authoritative.

Questions

About MCP

How do I install MCP?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is MCP safe to use with an AI agent?

Its trust score is 52 out of 100 (fair). It passes 3 of the 6 static security checks run so far; the failures are listed above. It has no ConduID identity yet, so agent calls to it are not receipted.

Is MCP still maintained?

The last commit was a year ago and 1 open issues. That's long enough that you should check whether the maintainer is responding to issues before depending on it.