What is MCP?
Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external data sources and tools. Think of it like a USB port for AI — plug in once, and it works everywhere.
Without MCP, your AI assistant has no idea what you've saved in Summify. With MCP, it can read your pods, search your content, and pull up that voice note you recorded on a walk last week — all in real time, right inside the conversation.
Summify's MCP server is a single URL. Add it to any compatible client and your entire knowledge base becomes available to your AI.
Why connect Summify?
Build on what you've read
Feed papers, talks, and lectures into a pod. Your AI cross-references findings, identifies gaps, and builds literature reviews grounded in your actual sources.
Learn from the best, create like them
Ingest every video from creators you admire. Your AI writes copy, plans campaigns, and builds strategies using their exact frameworks and language.
Your course material, always on call
Save every lecture and reading. Your AI tutor answers questions using your actual course material — not generic internet answers.
Connect the dots across sources
Store interview transcripts and source material in pods. Your AI finds connections across sources and pulls exact quotes for your story.
Real context while you code
Save technical talks, tutorials, and documentation. Your AI references real implementation details in your editor — not hallucinated examples.
Strategies backed by expertise
Build a pod of industry content — talks, podcasts, reports. Your AI builds strategies, competitor analyses, and plans grounded in expert knowledge.
What you need
An MCP client
Claude Desktop, Claude Code, Cursor, Windsurf, or any MCP-compatible tool.
Setup Guide
Two steps to connect your Summify library to any AI assistant. Just add the URL — your client handles the rest.
1Add to your MCP client
Choose your client below and add the configuration. You'll be prompted to sign in when you first connect.
Claude Desktop
Edit your claude_desktop_config.json file:
{
"mcpServers": {
"summify": {
"url": "https://summify.io/api/mcp"
}
}
}Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP client
Claude Code
Run this command in your terminal:
claude mcp add summify https://summify.io/api/mcpCursor
Add this to your .cursor/mcp.json file:
{
"mcpServers": {
"summify": {
"url": "https://summify.io/api/mcp"
}
}
}Windsurf
Add this to your Windsurf MCP configuration:
{
"mcpServers": {
"summify": {
"url": "https://summify.io/api/mcp"
}
}
}Any MCP Client
Use these details in any MCP-compatible client:
https://summify.io/api/mcpOAuth (automatic sign-in via browser)Streamable HTTP2Verify it works
Open your AI assistant and ask:
“List my Summify pods”
If it returns your pods, you're connected. That's it.
Alternative: API Key Setup
If your MCP client doesn't support OAuth, you can use an API key instead. Go to Account → API Access and generate a new key, then add a headers block to your config:
{
"mcpServers": {
"summify": {
"url": "https://summify.io/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Available Tools
Once connected, your AI assistant can use these tools to interact with your Summify library.
list_pods
List all your pods with content counts.
get_pod_overview
See all content titles, types, and topics in a pod at a glance.
get_pod_content
Get all content items in a pod, with optional type filtering.
search_content
Semantic search using AI embeddings — ask questions and get relevant passages without loading full transcripts.
batch_search_content
Search with multiple queries at once for broad research. Returns deduplicated results across all queries.
get_content_detail
Get the full summary and transcript for a specific content item.
Try these prompts
Once you're connected, try asking your AI assistant any of these:
“I have 50 Hormozi videos in my Marketing pod — build me a lead generation strategy”
“Cross-reference the key findings across all papers in my Research pod”
“What did Professor Smith say about thermodynamics in last week's lecture?”
“Find every mention of the CEO's revenue claims across my interview transcripts”
“Summarize the architectural patterns from the tech talks in my Engineering pod”
“Build a competitive analysis using the industry reports in my Strategy pod”