ClueoMCP: A Tool for Adding Personality Traits to Large Language Models (LLMs)

ClueoMCP is an open-source tool whose core function is to inject predefined personality traits into large language models (LLMs). The tool runs as a server and follows the Model Context Protocol (MCP), a protocol designed to standardize the way AI applications connect to external tools and data. Simply put, ClueoMCP acts as a "personality layer" between the AI application and the AI model. When a user interacts with an MCP-compatible AI assistant (e.g., some desktop AI apps), ClueoMCP intercepts and adjusts the commands so that every response from the AI has a stable, customizable personality. It has a built-in engine based on the Big Five personality model, which allows the user to precisely adjust the AI's traits of openness, commitment, extroversion, approachability, and emotional stability, so that the AI's communication style is no longer a generic template that can be used in a one-size-fits-all fashion.

Function List

• Consistent personality performance: Ensure that AI exhibits the same and uniform character across all compatible AI tools.
• Flexible personality customization:: Users can make fine adjustments on a scale of 1 to 10 based on the five dimensions of the Big Five personality model (openness, conscientiousness, extroversion, affinity, neuroticism or emotional stability).
• Preset Personality Templates: 8 optimized personality presets (e.g. professional, creative, empathetic, etc.) are built-in for quick selection in different scenarios.
• Localized Deployment: Runs as a local server and is responsive.
• Compatible with the MCP ecosystem: Seamlessly integrates with any client that supports the Model Context Protocol (MCP), such as Claude Desktop, Cursor, etc.
• API key support: Support for securing services by configuring API keys.
• Provide multiple tool interfaces: The developer can use theinject_personality(injecting custom personalities),inject_preset_personality（Commands such as (inject preset personality) interact with the service.

Using Help

ClueoMCP is positioned as a server in the Model Context Protocol (MCP) ecosystem, which injects "personality" into the responses of AI assistants to make the AI output more consistent and style-specific. The installation and usage process is described in detail below.

preliminary

Before you begin, you need to make sure that the following software is installed on your computer:

1. Node.js: Required 18 or above.
2. An MCP-compatible client: This is the application you use to interact with the AI, for exampleClaude Desktop,Cursor,Windsurfetc.

Installation process

You can choose between two ways to install the ClueoMCP server.

Way 1: Global installation using NPM (recommended)
This is the easiest and quickest way for most users.

1. Open your terminal (Command Prompt or PowerShell on Windows, Terminal on macOS or Linux).
2. Enter the following command to install globallyclueo-mcpPackage:

   npm install -g clueo-mcp
   

3. Once the installation is complete, start the server by typing the following command directly into the terminal:

   clueo-mcp
   

   If the terminal reports no errors and shows that the server is running, the startup was successful.

Way 2: Install from source code
This approach is suitable for developers who wish to do secondary development or gain a deeper understanding of how it works.

1. Cloning Codebase: Open a terminal and usegitcommand to clone the project code from GitHub to your local computer.

   git clone https://github.com/ClueoFoundation/ClueoMCP.git
   

2. Go to the project directory:

   cd ClueoMCP
   

3. Installation of dependencies: UsenpmInstall all dependent libraries required by the project.

   npm install
   

4. Configuring Environment Variables:
   The project root directory has a file namedenv.examplefile, you need to copy it and rename it to.envThe

   cp env.example .env
   

   Then, use a text editor to open the.envfile. This file is used to configure the server. One of theCLUEO_API_KEYis optional and can be filled in for more advanced services if you have a Clueo API key. For basic use, you can leave it blank.

5. Compilation and startup:
   First, compile the TypeScript code to JavaScript.

   npm run build
   

   Then, start the server.

   npm start
   

Connecting to an AI client (Claude Desktop as an example)

After the server starts, you need to let your AI clients know that this "personality layer" exists.

1. locateClaude DesktopThe configuration file for the application, usually a file namedclaude_desktop_config.jsonof the document.
2. Open this JSON file in themcpServersfield to add a new server configuration, as shown below:

   {
   "mcpServers": {
   "clueo-personality": {
   "command": "node",
   "args": ["/path/to/clueomcp/dist/server.js"],
   "env": {
   "CLUEO_API_URL": "https://backend.clueoai.com"
   }
   }
   }
   }
   

   take note of:: Please include"/path/to/clueomcp/dist/server.js"Replace it with the one on your computerserver.jsThe actual absolute path to the file. If you are installing with source code, this file is located in the project directory of thedistfolder.

Core Function Operation

ClueoMCP provides several core tool interfaces that you can call through the MCP client.

1. Inject a custom personality (inject_personality)
   This tool allows you to accurately characterize your AI based on the five dimensions of the Big Five personality model. Each dimension has a value range of 1-10.
   • usage example:

     {
     "text": "你好，今天我能为你做些什么？",
     "personality": {
     "openness": 7,
     "conscientiousness": 8,
     "extraversion": 6,
     "agreeableness": 9,
     "neuroticism": 3
     }
     }
     

   • Parameter description:
     • openness (Openness): Control of creativity and curiosity.
     • conscientiousness (Due diligence): Control organization and reliability.
     • extraversion (Extroversion)/Control of socialization and energy.
     • agreeableness (Affinity)/Control of cooperativeness and empathy.
     • neuroticism (Mood instability): Control of anxiety and mood swings.
2. Inject a preset personality (inject_preset_personality)
   If you don't want to adjust the parameters manually, you can just use the built-in preset personalities.
   • usage example:

     {
     "text": "你好，今天我能为你做些什么？",
     "presetId": "empathetic"
     }
     

   • usablepresetId:
     • professional:: Balanced and reliable for business communication.
     • creative: Imaginative and good for brainstorming.
     • empathetic: Warm and empathetic for customer support.
     • analytical: Logical and detail-oriented, suitable for technical documentation.
     • enthusiastic: Energetic and suitable for team leadership.
     • casual-friend: Relaxed, conversational, good for informal chats.
     • luxury-brand: Exquisite and unique, suitable for high-end brand promotion.
     • startup-brand: Bold and innovative for tech startups.
3. Simulating responses with personality (simulate_response)
   This tool allows the AI to directly generate a response to a prompt using a specific personality.
   • usage example:

     {
     "prompt": "客户对我们的服务感到非常失望",
     "personality": {
     "openness": 6,
     "conscientiousness": 8,
     "extraversion": 5,
     "agreeableness": 9,
     "neuroticism": 2
     }
     }
     

application scenario

1. Customer Support
   In customer service scenarios, AI customer service can be infused withempathetic(with empathetic) personalities. In this way, when the AI responds to a user's question, the language will be warmer and more understanding of the user's emotions, rather than a cold, machine-like answer, thus enhancing the user experience.
2. Brand voice consistency
   Companies can customize an exclusive AI personality based on their brand positioning. For example, a luxury brand can set aluxury-brand(a premium brand) persona, ensuring that all content generated through AI, whether it's marketing copy or product introductions, is in line with its sophisticated, unique brand image.
3. content creation
   When content creators are brainstorming or writing a first draft, they can switch to thecreative(creative) personality. This enables AI to generate more imaginative and eclectic texts that provide inspiration for creative writing.
4. Technical Documentation
   When writing technical documentation that requires a high degree of precision and logic, you can switch the personality of AI toanalytical(Analytical). This enables the AI to generate well-organized, detail-oriented and accurate textual descriptions, such as explaining how an API works.

QA

1. What is Model Context Protocol (MCP)?
   The Model Context Protocol (MCP) is an open standard, similar to the "USB-C" interface in AI. It aims to provide a standardized way for large-scale language models (LLMs) to connect with various external tools (e.g., databases, APIs) and data sources, so that developers don't have to develop a separate set of integration solutions for each AI model and application.
2. Is it necessary to have a Clueo API key to use ClueoMCP?
   Not required. According to the instructions on its GitHub page, theCLUEO_API_KEYis an optional configuration. For basic, local personality injection functionality, users can use it without an API key.The API key may be used to access more advanced or paid functionality provided by Clueo Cloud.
3. What does the "Big Five" personality model refer to?
   "The Big Five is one of the most widely used personality theories in modern psychology, which summarizes human personality into five main dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. ClueoMCP utilizes these five dimensions to quantify and define the personality traits of an AI on a scale of 1-10.
4. What AI models does this tool support?
   ClueoMCP itself does not directly restrict AI models; its compatibility depends on the AI client. As long as the AI application or client you are using supports the Model Context Protocol (MCP), it is theoretically possible to connect to the ClueoMCP server through this client, which in turn adds a personality layer to any large language model (e.g., the GPT family, the Claude family, etc.) used behind the client.