About AionMCP
AionMCP is a Go-based MCP server that imports OpenAPI, GraphQL, and AsyncAPI specifications and exposes their operations as tools. It is for developers connecting existing APIs to agents, with runtime tool reloading, self-learning, and generated documentation.
Key features
- 01Import three API specification types
- 02Expose API operations as agent tools
- 03Hot-reload tools at runtime
- 04Learn from execution patterns
- 05Update documentation and insights automatically
Use cases
- 01Make OpenAPI operations available to agents
- 02Connect GraphQL or AsyncAPI operations to agents
- 03Refresh tools without restarting the server
- 04Review execution insights and generated documentation
MCP support
As an MCP server, AionMCP exposes imported OpenAPI, GraphQL, and AsyncAPI operations as tools for MCP clients and agents.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEAionMCP - Autonomous Go MCP Server
AionMCP is an autonomous Go-based Model Context Protocol (MCP) server that dynamically imports OpenAPI, GraphQL, and AsyncAPI specifications and exposes them as tools to agents. It features self-learning capabilities, context-awareness, and autonomous documentation using Clean/Hexagonal architecture.
🌟 Key Differentiators
- Multi-Protocol Support: OpenAPI, GraphQL, and AsyncAPI specifications
- Autonomous Learning: Self-improving system that learns from execution patterns
- Dynamic Runtime: Hot-reloadable tools without service restart
- Clean Architecture: Maintainable, testable, and extensible design
- Auto-Documentation: Self-updating documentation and insights
📊 Project Status
Current Branch: main
Latest Commit: f4f41db
System Health: 99/100 (Excellent)
Active Tools: 3
Commits (7 days): 17
Status updated automatically
✨ Features
📦 What's Included
aionmcp/
├── bin/aionmcp-server # Main executable
├── cmd/server/ # Server entry point
├── internal/
│ ├── core/ # HTTP/gRPC servers & tool registry
│ ├── selflearn/ # Learning engine & BoltDB storage
│ └── autodocs/ # Documentation generation
├── pkg/
│ ├── importer/ # OpenAPI/GraphQL/AsyncAPI parsers
│ ├── agent/ # Agent registration API
│ └── feedback/ # Feedback collection
├── vscode-extension/ # VS Code extension
├── docs/ # Comprehensive documentation
├── examples/specs/ # Example API specifications
│ ├── petstore.yaml # OpenAPI 3.0 example
│ ├── blog.graphql # GraphQL example
│ └── user-events.yaml # AsyncAPI example
└── go.mod # Go dependencies
🚀 Quick Start
# Clone the repository
git clone https://github.com/kiransth77/aionmcp.git
cd aionmcp
# Build the server
go build -o bin/aionmcp cmd/server/main.go
# Run with default configuration
./bin/aionmcp
The server will start on http://localhost:8080 with learning enabled.
🏗️ Architecture
AionMCP follows Clean/Hexagonal Architecture principles:
┌─────────────────────────────────────────────────────────┐
│ Adapters Layer │
│ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │
│ │ HTTP │ │ gRPC │ │ Plugin │ │
│ │ Interface │ │ Interface │ │ Interface │ │
│ └─────────────┘ └─────────────┘ └─────────────┘ │
└─────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────┐
│ Core Layer │
│ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │
│ │ Tool │ │ Learning │ │ Auto │ │
│ │ Registry │ │ Engine │ │ Docs │ │
│ └─────────────┘ └─────────────┘ └─────────────┘ │
└─────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────┐
│ Infrastructure Layer │
│ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │
│ │ Storage │ │ Metrics │ │ Config │ │
│ │ (BoltDB) │ │(Prometheus) │ │ (Viper) │ │
│ └─────────────┘ └─────────────┘ └─────────────┘ │
└─────────────────────────────────────────────────────────┘
📈 Recent Activity
Recent Commits
f4f41db9fix: clean up TOON package formatting and remove duplicate package declaration (0h ago)8b97b173fix: update go-ci.yml to use Go 1.25 to match go.mod requirement (0h ago)ebeeee28fix: remove redundant newline in fmt.Println to pass linting (0h ago)a91eb5f4feat: add TOON (Token-Oriented Object Notation) for LLM context optimization (0h ago)bd39bb10fix: update Dockerfile Go version to 1.25 (0h ago)
Active Insights
📊 Total insights: 2
Activity updated automatically
⚡ Performance Statistics
| Metric | Value | Status |
|---|---|---|
| Success Rate | 97.0% | 🟢 Excellent |
| Avg Latency | 250.0ms | 🟡 Good |
| Total Executions | 42 | 📊 Tracking |
| Active Tools | 3 | 🔧 Running |
Statistics updated in real-time
📦 Installation
Prerequisites
- Go 1.21 or higher
- Git
From Source
git clone https://github.com/kiransth77/aionmcp.git
cd aionmcp
go mod download
go build -o bin/aionmcp cmd/server/main.go
📚 Usage
Basic Usage
# Start the server
./bin/aionmcp
# With custom configuration
./bin/aionmcp --config config.yaml
# Enable debug logging
AIONMCP_LOG_LEVEL=debug ./bin/aionmcp
API Endpoints
GET /api/v1/tools- List available toolsPOST /api/v1/tools/{tool}/execute- Execute a toolGET /api/v1/learning/stats- Learning statisticsGET /api/v1/learning/insights- System insights
🛠️ Development
Access: http://localhost:8080
### Docker
```bash
docker run -p 8080:8080 aionmcp:latest
Cloud (AWS/GCP/Azure)
Deploy the binary to your cloud provider, access via public URL.
Embedded
import "github.com/aionmcp/aionmcp/internal/core"
server := core.NewServer(logger)
server.Start()
🤝 Contributing
Active Insights
📊 Total insights: 2
Activity updated automatically
📄 License
📄 License
README last updated: 2025-11-27 22:47:18 AEST
This README is automatically updated with current project status and metrics.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.