The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
All In One Model Context Protocol
All-in-one MCP server with AI search, RAG, and multi-service integrations (GitLab/Jira/Confluence/YouTube) for AI-enhanced development workflows
Install
npx -y @smithery/cli install @nguyenvanduocit/all-in-one-model-context-protocol --client claudeInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →THE PROJECT IS MOVED
THE PROJECT HAS BEEN SPLIT AND MOVED TO INDIVIDUAL REPOSITORIES.
- Google Kit: Tools for Gmail, Google Calendar, Google Chat
- RAG Kit: Tools for RAG, Memory
- Dev Kit: Tools for developers, jira, confluence, gitlab, github, ...
- Fetch Kit: Tools for fetch, scrape, ...
- Research Kit: Tools for research, academic, reasoning, ...
MyMCP Server
A powerful Model Context Protocol (MCP) server implementation with integrations for GitLab, Jira, Confluence, YouTube, and more. This server provides AI-powered search capabilities and various utility tools for development workflows.
Community
For community support, discussions, and updates, please visit our forum at community.aiocean.io.
Prerequisites
- Go 1.23.2 or higher
- Various API keys and tokens for the services you want to use
Installation
Installing via Smithery
To install MyMCP Server for Claude Desktop automatically via Smithery (will guide you through interactive CLI setup):
npx -y @smithery/cli install @nguyenvanduocit/all-in-one-model-context-protocol --client claude
Note: Smithery will interactively prompt you for required configuration values and handle environment setup automatically
Installing via Go
- Install the server:
go install github.com/nguyenvanduocit/all-in-one-model-context-protocol@latest
- Manual setup required - Create a
.envfile with your configuration:
ENABLE_TOOLS=
QDRANT_HOST=
ATLASSIAN_HOST=
ATLASSIAN_EMAIL=
GITLAB_HOST=
GITLAB_TOKEN=
BRAVE_API_KEY=
ATLASSIAN_TOKEN=
GOOGLE_AI_API_KEY=
PROXY_URL=
OPENAI_API_KEY=
DEEPSEEK_API_KEY=
QDRANT_PORT=
GOOGLE_TOKEN_FILE=
GOOGLE_CREDENTIALS_FILE=
QDRANT_API_KEY=
- Config your claude's config:
{
"mcpServers": {
"my_mcp_server": {
"command": "all-in-one-model-context-protocol",
"args": ["-env", "/path/to/.env"],
}
}
}
Secret
GOOGLE_TOKEN_FILE is using for Google AI like Gemini, Google Search, Google Calendar, etc.
GOOGLE_CREDENTIALS_FILE is service account for advanced feature like Google Chat. You need to create project in Google Cloud Console and create service account. Then also need to create Google cloud bot chat for it. Ask any LLM model to create these for you.
Enable Tools
There are a hidden variable ENABLE_TOOLS in the environment variable. It is a comma separated list of tools group to enable. If not set, all tools will be enabled. Leave it empty to enable all tools.
Here is the list of tools group:
gemini: Gemini-powered searchfetch: Fetch toolsconfluence: Confluence toolsyoutube: YouTube toolsjira: Jira toolsgitlab: GitLab toolsscript: Script toolsrag: RAG toolsdeepseek: Deepseek AI tools
Available Tools
calendar_create_event
Create a new event in Google Calendar
Arguments:
Similar plugs
Reverse engineer anything with agents, from app behavior down to native binaries.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype…
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.