Fim One
Open-source agent platform for Global × China enterprises — wire every system through one agent core. Self-hosted, any LLM.
What is Fim One?
Fim One is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open-source agent platform for global × china enterprises — wire every system through one agent core. self-hosted, any llm.
Open-source agent platform for Global × China enterprises — wire every system through one agent core. Self-hosted, any LLM.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Open-source agent platform for Global × China enterprises —
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx fim-oneConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Fim One
FIM One is an open-source, self-hosted agent platform designed for Global and China enterprises that need to wire diverse backend systems through a single AI agent core. Built on FastAPI and Next.js, it supports any OpenAI-compatible LLM provider via LiteLLM (including OpenAI, DeepSeek, Anthropic, and Ollama) and provides RAG pipelines, code execution sandboxes, multi-database connectors, and MCP server integration. Teams adopt FIM One to deploy private, cost-controlled AI agents that work across Chinese enterprise databases (DM, KingbaseES) alongside standard PostgreSQL and MySQL stacks.
Prerequisites
- Docker and Docker Compose (recommended) or Python 3.11+ with uv and Node.js 18+ for local development
- An API key for at least one OpenAI-compatible LLM provider (OpenAI, DeepSeek, Anthropic, or a local Ollama instance)
- Optional: Jina API key for web tools and RAG embedding
- Git to clone the repository
- An MCP-compatible client if connecting via the MCP protocol
Clone the repository and configure environment
Clone the FIM One repo and copy the example environment file. Edit .env with your LLM provider credentials and optional settings.
git clone https://github.com/fim-ai/fim-one.git
cd fim-one
cp example.env .envSet required environment variables
Edit .env and set at minimum LLM_API_KEY and LLM_MODEL. Optionally set LLM_BASE_URL for non-OpenAI providers, JINA_API_KEY for web tools and RAG, and CODE_EXEC_BACKEND=docker for sandboxed code execution.
LLM_API_KEY=sk-your-key-here
LLM_MODEL=gpt-4o
LLM_BASE_URL=https://api.openai.com/v1
JINA_API_KEY=your-jina-key
CODE_EXEC_BACKEND=dockerStart with Docker Compose
Build and start all services using Docker Compose. The platform will be accessible on localhost once containers are running.
docker compose up --build -dConnect MCP servers
FIM One supports connecting external MCP servers in three ways: import an OpenAPI spec, use the AI chat builder, or connect via MCP server URL. Access this through the platform's agent configuration interface.
Configure agent workflows with DAG pipelines
Use the React Flow-based visual interface to define DAG-based multi-agent workflows, chaining tools such as document conversion, database queries, code execution, and RAG retrieval into automated pipelines.
Fim One Examples
Client configuration
Add FIM One as an MCP server in your Claude Desktop configuration after starting the platform locally.
{
"mcpServers": {
"fim-one": {
"command": "npx",
"args": ["fim-one"],
"env": {
"LLM_API_KEY": "sk-your-key-here",
"LLM_MODEL": "gpt-4o"
}
}
}
}Prompts to try
Use these prompts to explore FIM One's document processing, RAG, and database connectivity capabilities.
- "Convert this PDF to markdown and extract all tables"
- "Query the PostgreSQL database for sales data from last quarter and summarize trends"
- "Execute this Python script in the sandbox and return the output"
- "Search the knowledge base for documents about our Q3 roadmap"
- "Connect to the MySQL database at host:3306 and list all table names"Troubleshooting Fim One
Docker containers fail to start with networking errors
Ensure Docker Desktop is running and ports are not already in use. Run `docker compose logs` to identify the failing service. For Apple Silicon Macs, confirm Docker is using the correct platform (linux/amd64 emulation or arm64-native images).
LLM API calls return authentication errors
Verify that LLM_API_KEY is correctly set in your .env file with no trailing whitespace. For non-OpenAI providers, also set LLM_BASE_URL to the correct endpoint. Run `docker compose restart` after editing .env to pick up new values.
RAG retrieval returns no results
RAG requires a valid JINA_API_KEY for embedding generation. Confirm the key is set and that documents have been ingested through the platform's document import interface before querying the knowledge base.
Frequently Asked Questions about Fim One
What is Fim One?
Fim One is a Model Context Protocol (MCP) server that open-source agent platform for global × china enterprises — wire every system through one agent core. self-hosted, any llm. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Fim One?
Follow the installation instructions on the Fim One GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Fim One?
Fim One works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Fim One free to use?
Yes, Fim One is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
Fim One Alternatives — Similar Coding Agents Servers
Looking for alternatives to Fim One? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Fim One in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Fim One?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.