Skip to main content
Glama

AgentCraft MCP Server

by seyhunak
DOCUMENTATION.md2.05 kB
# AgentCraft MCP Server Documentation ## Project Overview AgentCraft is a partner product of Crafted™, providing an AI Agent framework for deploying intelligent agents that understand, learn, and evolve with business needs. The MCP server integrates with AgentCraft, enabling secure AI agent communication and data exchange. ## Key Features - Secure AI agent communication and data exchange - Supports premade and custom AI agents - Scalable and enterprise-ready architecture - Supports Windsurf MCP client - Example integrations for Cline and 5ire MCP clients ## System Architecture The system is built around the `AgentCraftServer` class which extends the MCP Server framework. Key components include: ### Core Components - **AgentCraftServer**: Main server implementation - Handles agent communication - Manages data processing pipelines - Implements security protocols - **DataProcessor**: Handles data transformation and processing - **AgentManager**: Manages agent lifecycle and interactions ### Dependencies - MCP server framework - spacy for NLP processing - aiohttp for HTTP communications - pyyaml for configuration management ## Installation & Setup ```bash # Installation commands will be added here ``` ## Configuration The system uses environment variables for configuration, managed through `.env` file and `python-dotenv`. Key configuration parameters: ### Required Environment Variables - `AGENTCRAFT_BEARER_TOKEN`: Authentication token for secure communications - `ENVIRONMENT`: Runtime environment (DEV/PROD) ### Logging Configuration - Log level: DEBUG - Format: `%(asctime)s - %(name)s - %(levelname)s - %(message)s` ### Environment-Specific Settings - DEV_BASE_URL: `127.0.0.1:8000` - PROD_BASE_URL: `api.we-crafted.com` ## API Documentation (To be populated after reviewing the source code) ## Data Flow (To be populated after reviewing the source code) ## Security Considerations (To be populated after reviewing the source code) ## Development Guidelines (To be populated after reviewing the source code)

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/seyhunak/agentcraft-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server