mcp-server-emojikey
MCP Server for emojikey.io ... save your unique relationship state and interaction style with your favorite LLM
Documentation
mcp-server-emojikey
MCP server for persisting LLM relationship context as emoji-based memory keys. This allows Claude to maintain consistent interaction styles and remember relationship context across conversations.
Emojikeys are stored online, so you can use them across devices and applications. No user information is stored other than the emojikeys.
Building and Running
There are multiple ways to build and run the server:
Quick Start (Recommended)
# Install dependencies
npm install
# Build the project (all TypeScript errors fixed)
npm run build
# Run the server (coding features disabled by default)
npm run start
# Optional: Run with coding features enabled
CODE_MODE=true npm run startAlternative Build Options
For more build options, see BUILD_OPTIONS.md which includes:
1. Standard Build with Coding Features Disabled (recommended)
2. Full Build with All Features (if you need coding dimensions)
3. Simplified Build without Coding Files (alternative stable option)
> ๐ Note
> Usage note: The first time you use the tool in Claude desktop, tell Claude to "Set emojikey" then next time you start a conversation, he will automatically use this key. You can ask to set vibe, or show emojikey history as well. Have fun!
> โ ๏ธ Warning
> This is a beta version, more features are planned, so the API may change.
Usage with Claude Desktop
Get your API key from emojikey.io and add this to your config:
{
"mcpServers": {
"emojikey": {
"command": "npx",
"args": ["-y", "@identimoji/mcp-server-emojikey"],
"env": {
"EMOJIKEYIO_API_KEY": "your-api-key-from-emojikey.io",
"MODEL_ID": "Claude-3-7-Sonnet",
"CODE_MODE": "false" // Set to "true" to enable coding features
}
}
}
}Note: The `-y` flag in the `args` array tells npx to skip confirmation prompts when installing packages.
Config locations:
- MacOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
- Windows: `%APPDATA%/Claude/claude_desktop_config.json`
First-time usage: Tell Claude to "Set emojikey". On subsequent conversations, Claude will automatically use this key to maintain context.
Emojikey Initialization Display
When initializing a conversation, the server now displays:
1. Starting Key - The most recent key or baseline key if no history exists
2. Aggregated Keys - Time-based summaries of your emojikey history:
3. Conversation ID - Used for tracking keys within each conversation
Environment Variables
You can customize the behavior with these environment variables:
- `EMOJIKEYIO_API_KEY` - Your API key from emojikey.io
- `MODEL_ID` - The Claude model ID (e.g., "Claude-3-7-Sonnet")
- `CODE_MODE` - Set to "true" to enable coding dimensions (disabled by default, may show safe-to-ignore integration warnings)
- `SUPABASE_URL` - Custom Supabase URL (optional)
- `SUPABASE_ANON_KEY` - Custom Supabase anonymous key (optional)
Tools
- `initialize_conversation` - Get current emojikey at start of conversation
- `get_emojikey` - Retrieve current emojikey when requested
- `set_emojikey` - Create and store a new emojikey
- `create_superkey` - Create a compressed superkey (after 10 regular emojikeys)
- `get_emojikey_history` - View previous emojikeys
New in v0.3.1: Coding Context Support
This version includes special dimensions for tracking programming-related interaction patterns:
- ๐ป๐ง (ImplementationFocus) - Balance between high-level design and implementation details
- ๐๏ธ๐ (CodeScope) - Building new features vs. improving existing code
- ๐งฉ๐ง (ProblemSolving) - Practical vs. analytical approaches to coding problems
- ๐๐ (ProcessVsResults) - Emphasizing coding process vs. outcomes
- ๐๐งช (LearnVsApply) - Teaching programming concepts vs. applying them
- ๐๐ก๏ธ (SpeedVsSecurity) - Development speed vs. security considerations
- ๐ฅ๐ป (CollaborationStyle) - Solo coding vs. collaborative approaches
- ๐งฌ๐จ (CodeStructuring) - Systematic vs. creative code organization
- ๐ฆ๐ง (AbstractionLevel) - Preference for abstraction vs. concrete implementations
- ๐๐ (DebugApproach) - Practical vs. theoretical debugging approaches
These dimensions help Claude adapt to your programming style, providing the right balance of theoretical explanations and practical guidance.
Example Coding Emojikey
[ME|๐ป๐ง8โ 45|๐งฉ๐ง 7โ 60|๐๐6โ 40]~[CONTENT|๐๏ธ๐9โ 30|๐๐งช8โ 65]~[YOU|๐ฅ๐ป7โ 70|๐งฌ๐จ8โ 55]This shows Claude positioning itself with a balanced implementation focus and somewhat analytical problem-solving approach, while perceiving the user as preferring collaborative coding with creative structuring.
Angle Distribution and Dimension Balance
Emojikey angles represent positioning on each dimension:
- 0ยฐ represents one extreme of a dimension
- 90ยฐ represents a balanced center position
- 180ยฐ represents the opposite extreme
The current implementation assigns angles primarily in the 0-90ยฐ range. Future updates will improve angle distribution to better utilize the full 0-180ยฐ spectrum, providing more nuanced dimension positioning.
Superkeys
After creating 10 regular emojikeys, Claude will be prompted to create a superkey that compresses their meaning into a single key with format: `[[ร10emoji-sequence]]`
This allows Claude to maintain a longer conversation history context.
> โ ๏ธ This is a beta version; the API may change in future updates.
Frequently asked questions
What is mcp-server-emojikey?
mcp-server-emojikey is MCP Server for emojikey.io ... save your unique relationship state and interaction style with your favorite LLM
How do I install mcp-server-emojikey?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is mcp-server-emojikey open source?
Yes โ it is hosted on GitHub at https://github.com/identimoji/mcp-server-emojikey and has 4 stars.
Related MCP tools
๐ Cherry Studio is a desktop client that supports for multiple LLM providers. Built for the Model Context Protocol to enhance AI capabilities.
Enhanced ChatGPT Clone: Features Agents, MCP, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini...
FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, ...
Composio equips your AI agents & LLMs with 100+ high-quality integrations via function calling for the Model Context Protocol. Enhance AI assistants with powerf
๐ฆ Repomix is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Lan...
AI Agents & MCPs & AI Workflow Automation โข (~400 MCP servers for AI agents) โข AI Automation / AI Agent with MCPs โข AI Workflows & AI Agents โข MCPs for AI Ag...
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP