I craft unique cereal names, stories, and ridiculously cute Cereal Baby images.

eliza-plugin-mcp
ElizaOS plugin allowing agents to connect to MCP servers
3 years
Works with Finder
4
Github Watches
5
Github Forks
19
Github Stars
MCP Plugin for ElizaOS
This plugin integrates the Model Context Protocol (MCP) with ElizaOS, allowing agents to connect to multiple MCP servers and use their resources, prompts, and tools.
🔍 What is MCP?
The Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. It provides a standardized way to connect LLMs with the context they need.
This plugin allows your ElizaOS agents to access multiple MCP servers simultaneously, each providing different capabilities:
- Resources: Context and data for the agent to reference
- Prompts: Templated messages and workflows
- Tools: Functions for the agent to execute
📦 Installation
Install the plugin in your ElizaOS project:
- npm
npm install @fleek-platform/eliza-plugin-mcp
- pnpm
pnpm install @fleek-platform/eliza-plugin-mcp
- yarn
yarn add @fleek-platform/eliza-plugin-mcp
- bun
bun add @fleek-platform/eliza-plugin-mcp
🚀 Usage
- Add the plugin to your character configuration:
{
"name": "Your Character",
"plugins": ["@fleek-platform/eliza-plugin-mcp"],
"settings": {
"mcp": {
"servers": {
"github": {
"type": "stdio",
"name": "Code Server",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-github"]
}
}
}
}
}
⚙️ Configuration Options
MCP supports two types of servers: "stdio" and "sse". Each type has its own configuration options.
Common Options
Option | Type | Description |
---|---|---|
type |
string | The type of MCP server: "stdio" or "sse" |
name |
string | The display name of the server |
timeout |
number | Timeout in seconds for tool calls (default: 60) |
disabled |
boolean | Whether the server is disabled |
stdio Server Options
Option | Type | Description |
---|---|---|
command |
string | The command to run the MCP server |
args |
string[] | Command-line arguments for the server |
env |
object | Environment variables to pass to the server |
cwd |
string | Optional Working directory to run the server in |
sse Server Options
Option | Type | Description |
---|---|---|
url |
string | The URL of the SSE endpoint |
🛠️ Using MCP Capabilities
Once configured, the plugin automatically exposes MCP servers' capabilities to your agent:
Context Providers
The plugin includes three providers that add MCP capabilities to the agent's context:
-
MCP_SERVERS
: Lists available servers and their tools, resources and prompts
🔄 Plugin Flow
The following diagram illustrates the MCP plugin's flow for tool selection and execution:
graph TD
%% Starting point - User request
start[User Request] --> action[CALL_TOOL Action]
%% MCP Server Validation
action --> check{MCP Servers Available?}
check -->|No| fail[Return No Tools Available]
%% Tool Selection Flow
check -->|Yes| state[Get MCP Provider Data]
state --> prompt[Create Tool Selection Prompt]
%% First Model Use - Tool Selection
prompt --> model1[Use Language Model for Tool Selection]
model1 --> parse[Parse Selection]
parse --> retry{Valid Selection?}
%% Second Model Use - Retry Selection
retry -->|No| feedback[Generate Feedback]
feedback --> model2[Use Language Model for Retry]
model2 --> parse
%% Tool Selection Result
retry -->|Yes| toolAvailable{Tool Available?}
toolAvailable -->|No| fallback[Fallback Response]
%% Tool Execution Flow
toolAvailable -->|Yes| callTool[Call MCP Tool]
callTool --> processResult[Process Tool Result]
%% Memory Creation
processResult --> createMemory[Create Memory Record]
createMemory --> reasoningPrompt[Create Reasoning Prompt]
%% Third Model Use - Response Generation
reasoningPrompt --> model3[Use Language Model for Response]
model3 --> respondToUser[Send Response to User]
%% Styling
classDef model fill:#f9f,stroke:#333,stroke-width:2px;
classDef decision fill:#bbf,stroke:#333,stroke-width:2px;
classDef output fill:#bfb,stroke:#333,stroke-width:2px;
class model1,model2,model3 model;
class check,retry,toolAvailable decision;
class respondToUser,fallback output;
📋 Example: Setting Up Multiple MCP Servers
Here's a complete example configuration with multiple MCP servers of both types:
{
"name": "Developer Assistant",
"plugins": ["@elizaos/plugin-mcp", "other-plugins"],
"settings": {
"mcp": {
"servers": {
"github": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-github"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
}
},
"puppeteer": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-puppeteer"]
},
"google-maps": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-google-maps"],
"env": {
"GOOGLE_MAPS_API_KEY": "<YOUR_API_KEY>"
}
}
},
"maxRetries": 2
}
}
}
🔒 Security Considerations
Please be aware that MCP servers can execute arbitrary code, so only connect to servers you trust.
🔍 Troubleshooting
If you encounter issues with the MCP plugin:
- Check that your MCP servers are correctly configured and running
- Ensure the commands are accessible in the ElizaOS environment
- Review the logs for connection errors
- Verify that the plugin is properly loaded in your character configuration
👥 Contributing
Thanks for considering contributing to our project!
How to Contribute
- Fork the repository.
- Create a new branch:
git checkout -b feature-branch-name
. - Make your changes.
- Commit your changes using conventional commits.
- Push to your fork and submit a pull request.
Commit Guidelines
We use Conventional Commits for our commit messages:
-
test
: 💍 Adding missing tests -
feat
: 🎸 A new feature -
fix
: 🐛 A bug fix -
chore
: 🤖 Build process or auxiliary tool changes -
docs
: ✏️ Documentation only changes -
refactor
: 💡 A code change that neither fixes a bug or adds a feature -
style
: 💄 Markup, white-space, formatting, missing semi-colons...
📄 License
This plugin is released under the same license as ElizaOS.
相关推荐
Converts Figma frames into front-end code for various mobile frameworks.
I find academic articles and books for research and literature reviews.
A unified API gateway for integrating multiple etherscan-like blockchain explorer APIs with Model Context Protocol (MCP) support for AI assistants.
Mirror ofhttps://github.com/suhail-ak-s/mcp-typesense-server
本项目是一个钉钉MCP(Message Connector Protocol)服务,提供了与钉钉企业应用交互的API接口。项目基于Go语言开发,支持员工信息查询和消息发送等功能。
Short and sweet example MCP server / client implementation for Tools, Resources and Prompts.
Reviews

user_dnQKd8H9
I recently started using the eliza-plugin-mcp by fleek-platform, and I'm thoroughly impressed. The integration was seamless, and its functionality has significantly enhanced my MCP applications. The documentation is clear, and the support from the developers is outstanding. I highly recommend this plugin to anyone looking to boost their MCP system’s capabilities. You can check it out here: https://github.com/fleek-platform/eliza-plugin-mcp.