Skip to main content
Glama

Civitai MCP Server

Civitai MCP Server

A Model Context Protocol (MCP) server that provides AI assistants with comprehensive access to Civitai's vast collection of AI models, creators, and generated content. Browse, search, and discover AI models seamlessly through your favorite MCP-compatible AI assistant.

Features

🔍 Model Discovery

  • Advanced Search: Find AI models with flexible filtering by type, creator, tags, and more
  • Browse Categories: Explore models by type (Checkpoints, LoRA, ControlNet, etc.)
  • Popular & Trending: Discover the most downloaded and highest-rated models
  • Latest Models: Stay up-to-date with newly uploaded models
  • Hash Lookup: Find models by file hash for verification

👨‍💻 Creator & Community

  • Creator Profiles: Browse and search for model creators
  • Creator Collections: View all models from specific creators
  • Tag System: Explore models through Civitai's comprehensive tagging system

🖼️ Generated Content

  • Image Gallery: Browse AI-generated images with detailed metadata
  • Generation Parameters: Access prompt, settings, and model information
  • Community Showcase: Discover inspiring creations from the community

📊 Model Intelligence

  • Detailed Model Info: Complete model specifications, versions, and files
  • Version History: Track model updates and improvements
  • Download URLs: Direct access to model downloads with authentication support
  • Content Safety: Access scan results for pickle and virus safety

Installation

Prerequisites

  • Node.js 18+
  • npm or yarn
  • A Civitai API key (optional, but recommended for higher rate limits)

Quick Start

  1. Clone the repository:
git clone https://github.com/Cicatriiz/civitai-mcp-server.git cd civitai-mcp-server
  1. Install dependencies:
npm install
  1. Build the server:
npm run build
  1. Set up your API key (optional):
export CIVITAI_API_KEY="your_api_key_here"
  1. Run the server:
npm start

Getting a Civitai API Key

  1. Visit Civitai API Keys
  2. Log in to your Civitai account
  3. Generate a new API key
  4. Copy the key and set it as an environment variable

Configuration

MCP Client Setup

Add the server to your MCP client configuration:

Claude Desktop

Add to your claude_desktop_config.json:

{ "mcpServers": { "civitai": { "command": "node", "args": ["/path/to/civitai-mcp-server/dist/index.js"], "env": { "CIVITAI_API_KEY": "your_api_key_here" } } } }
Other MCP Clients

Configure according to your client's documentation, using:

  • Command: node /path/to/civitai-mcp-server/dist/index.js
  • Transport: stdio
  • Environment: CIVITAI_API_KEY=your_key

Usage Examples

Search for SDXL LoRA models related to anime: - Tool: search_models - Query: "anime" - Types: ["LORA"] - BaseModels: ["SDXL 1.0"]
Get the most downloaded models this week: - Tool: get_popular_models - Period: "Week" - Limit: 10

Model Details

Get comprehensive information about a specific model: - Tool: get_model - ModelId: 12345

Browse Generated Images

Explore recent AI-generated images: - Tool: browse_images - Sort: "Newest" - Limit: 50 - NSFW: "None"

Available Tools

ToolDescriptionKey Parameters
search_modelsSearch models with filtersquery, types, sort, baseModels
get_modelGet detailed model infomodelId
get_model_versionGet version detailsmodelVersionId
get_model_version_by_hashFind model by file hashhash
browse_imagesBrowse generated imagessort, period, modelId
get_creatorsSearch creatorsquery, limit
get_tagsBrowse model tagsquery, limit
get_popular_modelsMost popular modelsperiod, limit
get_latest_modelsNewest modelslimit
get_top_rated_modelsHighest rated modelsperiod, limit
search_models_by_tagModels with specific tagtag, sort
search_models_by_creatorModels by creatorusername, sort
get_models_by_typeFilter by model typetype, sort
get_download_urlGet model download URLmodelVersionId

API Reference

Model Types

  • Checkpoint - Full Stable Diffusion models
  • LORA - Low-Rank Adaptation models
  • TextualInversion - Embedding models
  • Hypernetwork - Hypernetwork models
  • ControlNet - ControlNet models
  • AestheticGradient - Aesthetic gradient models
  • Poses - Pose models

Sort Options

  • Highest Rated - Best community ratings
  • Most Downloaded - Most popular downloads
  • Newest - Recently uploaded

Time Periods

  • AllTime - No time restriction
  • Year - Past 12 months
  • Month - Past 30 days
  • Week - Past 7 days
  • Day - Past 24 hours

API Coverage

This MCP server implements all major Civitai API v1 endpoints:

  • /api/v1/models - List and search models
  • /api/v1/models/:id - Get specific model
  • /api/v1/model-versions/:id - Get model version
  • /api/v1/model-versions/by-hash/:hash - Get version by hash
  • /api/v1/images - Browse images
  • /api/v1/creators - List creators
  • /api/v1/tags - List tags
  • ✅ Download URLs with authentication support

Model Types Supported

  • Checkpoint: Full Stable Diffusion models
  • LORA: Low-Rank Adaptation models
  • TextualInversion: Textual inversion embeddings
  • Hypernetwork: Hypernetwork models
  • AestheticGradient: Aesthetic gradient models
  • Controlnet: ControlNet models
  • Poses: Pose models

Content Filtering

The server supports Civitai's content filtering system:

  • NSFW content levels (None, Soft, Mature, X)
  • Commercial use permissions
  • Model licensing options

Error Handling

The server includes comprehensive error handling for:

  • API rate limiting
  • Network connectivity issues
  • Invalid parameters
  • Authentication errors
  • Data validation

Development

Project Structure

civitai-mcp-server/ ├── src/ │ ├── index.ts # Main server implementation │ ├── civitai-client.ts # Civitai API client │ └── types.ts # TypeScript type definitions ├── dist/ # Compiled JavaScript output ├── tests/ # Test files └── docs/ # Additional documentation

Building from Source

# Install dependencies npm install # Build TypeScript npm run build # Run tests npm test # Development mode with hot reload npm run dev

Testing

# Run comprehensive test suite npm test # Test specific endpoints node test.js node comprehensive-test.js

Contributing

We welcome contributions! Please see our Contributing Guide for details.

Development Setup

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Add tests for new functionality
  5. Submit a pull request

Limitations

  • Rate Limiting: Civitai API has rate limits. Use an API key for higher limits.
  • NSFW Content: Some content may be filtered based on your account settings.
  • Model Availability: Some models may be temporarily unavailable or require authentication.

Troubleshooting

Common Issues

Server won't start:

  • Ensure Node.js 18+ is installed
  • Check that all dependencies are installed (npm install)
  • Verify the build completed successfully (npm run build)

API rate limiting:

  • Get a Civitai API key and set the CIVITAI_API_KEY environment variable
  • Reduce request frequency if experiencing limits

Models not found:

  • Check model ID accuracy
  • Verify the model hasn't been removed or made private
  • Ensure proper spelling in search queries

License

MIT License - see LICENSE file for details

Support

-
security - not tested
A
license - permissive license
-
quality - not tested

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

Provides AI assistants with access to Civitai's collection of AI models, enabling users to browse, search, and discover AI models through MCP-compatible AI assistants.

  1. Features
    1. 🔍 Model Discovery
    2. 👨‍💻 Creator & Community
    3. 🖼️ Generated Content
    4. 📊 Model Intelligence
  2. Installation
    1. Prerequisites
    2. Quick Start
    3. Getting a Civitai API Key
  3. Configuration
    1. MCP Client Setup
  4. Usage Examples
    1. Basic Model Search
    2. Find Popular Models
    3. Model Details
    4. Browse Generated Images
  5. Available Tools
    1. API Reference
      1. Model Types
      2. Sort Options
      3. Time Periods
    2. API Coverage
      1. Model Types Supported
        1. Content Filtering
          1. Error Handling
            1. Development
              1. Project Structure
              2. Building from Source
              3. Testing
            2. Contributing
              1. Development Setup
            3. Limitations
              1. Troubleshooting
                1. Common Issues
              2. License
                1. Related Projects
                  1. Support

                    Related MCP Servers

                    • -
                      security
                      A
                      license
                      -
                      quality
                      🔍 Enable AI assistants to search, access, and analyze PubMed articles through a simple MCP interface.
                      Last updated -
                      42
                      Python
                      MIT License
                      • Apple
                      • Linux
                    • -
                      security
                      A
                      license
                      -
                      quality
                      Enables AI assistants to interact with Mealie recipe databases through MCP clients like Claude Desktop.
                      Last updated -
                      7
                      Python
                      MIT License
                    • A
                      security
                      A
                      license
                      A
                      quality
                      Connects AI assistants to external data sources (Google, Bing, etc.) via SearchAPI.site, implementing the Model Context Protocol (MCP) for secure and contextual access to web information.
                      Last updated -
                      1
                      1
                      8
                      TypeScript
                      MIT License
                      • Linux
                      • Apple
                    • A
                      security
                      A
                      license
                      A
                      quality
                      Enables AI assistants to discover, retrieve details about, and manage MCP (Model Context Protocol) servers that provide additional tools and capabilities on demand.
                      Last updated -
                      4
                      345
                      3
                      JavaScript
                      AGPL 3.0
                      • Linux
                      • Apple

                    View all related MCP servers

                    MCP directory API

                    We provide all the information about MCP servers via our MCP API.

                    curl -X GET 'https://glama.ai/api/mcp/v1/servers/Cicatriiz/civitai-mcp-server'

                    If you have feedback or need assistance with the MCP directory API, please join our Discord server