🤖 MCP Integration
Connect StoryLab with Claude Code, ChatGPT, and other AI assistants
What is MCP?
Model Context Protocol (MCP) lets you connect StoryLab directly to AI assistants like Claude Code, ChatGPT, and Gemini. This enables:
- 🎯 Direct Content Analysis - Analyze brand stories from within your AI workflow
- 📊 Automated Query Generation - Generate search queries for content discovery
- 🔍 Semantic Search - Search your content library programmatically
- 📁 Document Processing - Analyze PDFs and images via chat interface
- 📚 Story Management - Create, edit, and organize stories through AI commands
Claude Code ✅ Working
Integrate StoryLab directly into your Claude Code workflow for seamless content analysis.
Choose Your Configuration
Setup Instructions
Add this to your claude_desktop_config.json file:
- 📱 Find your config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
- macOS:
- 📝 Copy the JSON configuration from the box above
- 📁 Paste it into your config file (create the file if it doesn't exist)
- ✏️ Update your credentials in the configuration
- 🔄 Restart Claude Desktop to load the integration
ChatGPT 🚧 Beta
Use StoryLab with ChatGPT through OpenAI's GPT actions or custom API integration.
GPT Action Configuration
{
"openapi": "3.1.0",
"info": {
"title": "StoryLab MCP API",
"version": "1.0.0"
},
"servers": [
{
"url": "https://storylab.scope3.com/mcp"
}
],
"paths": {
"/api/analyze-content": {
"post": {
"operationId": "analyzeContent",
"summary": "Analyze marketing content for brand stories",
"requestBody": {
"required": true,
"content": {
"application/json": {
"schema": {
"type": "object",
"properties": {
"content": {"type": "string"},
"language": {"type": "string", "default": "English"}
}
}
}
}
}
}
}
},
"components": {
"securitySchemes": {
"BearerAuth": {
"type": "http",
"scheme": "bearer"
}
}
}
}
Google Gemini 📋 Planned
Integration with Google Gemini through function calling and extensions (coming soon).
Function Calling Schema
{
"name": "storylab_analyze_content",
"description": "Analyze marketing content to extract brand stories and narratives",
"parameters": {
"type": "object",
"properties": {
"content": {
"type": "string",
"description": "Marketing content or brief to analyze"
},
"language": {
"type": "string",
"description": "Target language for analysis",
"default": "English"
}
},
"required": ["content"]
}
}
🧪 Test Your Integration
Once configured, try these sample queries with your AI assistant. You can also use the built-in sample prompts that appear in your MCP integration!
🎯 Brand Story Analysis:
"Analyze this marketing content: 'Our eco-friendly brand focuses on sustainable materials and ethical manufacturing...'"
"Analyze this marketing content: 'Our eco-friendly brand focuses on sustainable materials and ethical manufacturing...'"
🔍 Content Search:
"Search for articles about sustainable fashion trends"
"Search for articles about sustainable fashion trends"
📋 Query Generation:
"Based on my brand analysis, generate search queries to find similar brands and competitors"
"Based on my brand analysis, generate search queries to find similar brands and competitors"
🏭 Complete Research Workflow:
"Do a comprehensive brand research analysis for [your brand content]"
"Do a comprehensive brand research analysis for [your brand content]"
💡 Try Sample Prompts:
Look for "Sample Prompts" in your MCP integration - these include guided workflows like "Analyze Marketing Content" and "Complete Brand Research"
Look for "Sample Prompts" in your MCP integration - these include guided workflows like "Analyze Marketing Content" and "Complete Brand Research"
📈 Get analytics:
"Show me my StoryLab usage statistics and recent activity"
"Show me my StoryLab usage statistics and recent activity"
📚 Story Management:
"List all my stories, then edit the one about sustainability to add a new article from URL: https://example.com/green-tech"
"List all my stories, then edit the one about sustainability to add a new article from URL: https://example.com/green-tech"
🔍 Find Similar Content:
"Find stories from my initial prompt: 'AI-powered healthcare innovation' and show me what's already in my collection"
"Find stories from my initial prompt: 'AI-powered healthcare innovation' and show me what's already in my collection"
🚀 Complete Story Workflow:
"Use guided story creation to help me build a comprehensive story about renewable energy trends with a brief first"
"Use guided story creation to help me build a comprehensive story about renewable energy trends with a brief first"
🔧 Available MCP Tools
StoryLab provides these tools through the MCP protocol:
- analyze_content - Extract brand stories from marketing content
- generate_queries - Generate search queries based on brand analysis
- search_content - Perform semantic search across your content
- list_my_stories - View all your existing brand stories
- add_new_story - Create new stories with streamlined 3-step workflow (requires confirmation)
- edit_story - Update story name, description, and manage articles
- delete_story - Permanently delete stories (with confirmation)
- manage_story_artifact - Add or remove articles from existing stories
- find_closest_stories - Find similar stories for any URL
- find_stories_from_prompt - Search existing stories based on initial prompts
- guided_story_creation - Complete guided workflow for creating stories with brief generation (requires confirmation)
⚡ Quick Start
- Choose Your Setup: Select your Claude version and authentication method above
- Copy Configuration: Click "Copy Configuration" to get your personalized setup
- Add to Claude: Paste the configuration into your Claude settings
- Update Credentials: Replace placeholder email/password with your actual StoryLab credentials
- Test & Explore: Try sample queries like "What brand stories do I have live in StoryLab?"
That's it! No tokens needed - just use your regular StoryLab login credentials in the configuration.