Direct answer: Edithly's MCP Server makes your entire document knowledge base available to Claude, Cursor, and other MCP-compatible AI agents as a live, searchable context source. Your documents become part of the AI's working memory.
What Is the Model Context Protocol?
The Model Context Protocol (MCP) is an open standard introduced by Anthropic that defines how AI models access external tools and data. Think of it as USB-C for AI integrations — a single, standardised connector that any compliant AI tool can use to access any compliant data source.
Before MCP, connecting your documents to an AI required custom engineering for each tool — different APIs, different authentication flows, different data formats. MCP standardises this: one integration, any tool.
Edithly's MCP Server implements this protocol, exposing your uploaded documents, repositories, and generated visuals as a live resource for AI agents.
What Becomes Possible with Edithly MCP
Claude + Your Documents
Instead of pasting document sections into Claude's chat, connect Edithly's MCP Server to Claude Desktop. Now ask Claude: "What does our Q3 strategy document say about headcount?" Claude retrieves the answer directly from your Edithly repository — cited and accurate.
Cursor + Your Codebase Docs
Developers using Cursor for coding can connect Edithly's MCP Server to their Cursor instance. Technical specifications, API documentation, and architecture decision records stored in Edithly become accessible to Cursor as context — improving code generation accuracy significantly.
Custom AI Agents + Your Knowledge Base
Building an internal AI assistant for your team? Connect it to Edithly's MCP Server and it gains access to your entire organisational knowledge base — policies, processes, product documentation, research — without you having to rebuild the document management layer.
Setup in Under 5 Minutes
Add the Edithly MCP Server to your Claude Desktop configuration:
{
"mcpServers": {
"edithly": {
"command": "npx",
"args": ["-y", "@edithly/mcp-server"],
"env": {
"EDITHLY_API_KEY": "your-api-key-here"
}
}
}
}
Save the config. Restart Claude Desktop. Your Edithly repositories are now available as context sources.
MCP Server Capabilities
Edithly's MCP Server exposes the following tools to connected AI agents:
- search_repository — semantic search across your document collections
- get_document_section — retrieve a specific section from a named document
- list_repositories — list available knowledge bases
- generate_visual — trigger visual generation (mind map, flashcard set, etc.) from an AI workflow
- get_chat_history — retrieve previous Q&A sessions for context
Use Cases by Role
Developers and Technical Teams
Connect Edithly to Cursor or a custom coding assistant. Technical specifications, API documentation, and internal runbooks stored in Edithly become live context — reducing hallucinations and improving the accuracy of AI-generated code.
Research Teams
Connect a research paper repository to Claude. Ask synthesis questions across hundreds of papers without manually searching each one. Claude cites specific papers in its responses.
Legal and Compliance
Connect contract repositories to a legal AI assistant. Query across contracts, flag specific clauses, and get sourced answers — without manual document search.
Product and Strategy Teams
Connect product documentation, competitor research, and strategy documents to your AI tooling. Every AI-generated output — strategy drafts, competitive analyses, product requirement documents — is grounded in your actual organisational knowledge.
Why MCP Over Direct API Integration
Direct API integrations require custom code for each AI tool. When the AI tool updates its API, your integration breaks. When you want to use a different AI tool, you rebuild.
MCP is a standard. Build the integration once — Edithly's MCP Server — and it works with every MCP-compatible tool today and in the future. Lower maintenance, higher interoperability.
Start Using Edithly MCP
Connect your documents to Claude and other AI tools in under 5 minutes. Available on Pro and API plans.
