MCP Clients
kibitz
No more being stuck with whatever tools your chat app decided to include.
Needs: gitnpm
gemini-mcp-desktop-client
No more living in a terminal window just to use your MCP tools.
Needs:
opencode-x
Your terminal coding assistant can run on any model you already pay for.
Needs:
oryxos
A self-hosted home for AI agents that plug into MCP tools you already run.
Needs:
zin-mcp-client
A bridge between MCP servers and the AI models running on your own computer.
Needs:
extra
A small engine that gives your app focused assistants wired to your own MCP tools.
Needs:
gemini-desktop
Turns Google's Gemini into a desktop chat window that can reach your tools.
Needs: Python 3.13+ (as specified in pyproject.toml)uv (Modern Python package installer and resolver)
agentic-os
Brings your goals, backlog, and daily plan into whichever coding agent you use.
Needs:
mcp-client-nodejs
Lets you chat with any local tool server from your own terminal window.
Needs:
dsh-vscode-layout
Transform DeepSeek Harness web into a VS Code style IDE with tabs and file tree.
Needs:
bench-studio-public
Brings a local creative studio with 37 AI model routes into MCP-compatible apps.
Needs:
xiaozhi-ios
Chat with a Live2D AI assistant using voice and multimodal features on iOS.
Needs: Xcode 16 or laterAn iOS 17.6+ device or simulatorAn Apple Developer account (for device signing)
openbot
OpenBot client — run and manage AI coworkers with their own browsers and files
Needs: Docker, for PostgreSQL, browser computers, the supervisor, and the shipped Bots.Bun 1.3+, for the app and API server.A CopilotKit Intelligence project and license.A model key. The proof-of-concept Bot uses OpenAI; the LangGraph Bot can use OpenAI, Anthropic, or Google.
multi-agent
Brings multi-agent travel assistance into your chat or command line app.
Needs:
trade-agent-mcp
Skip manual trading steps by controlling stocks, crypto, and options via AI.
Needs: