Cline Integration

You have a complex code repository with tangled dependencies, which makes understanding relationships challenging. You want to generate a clear, detailed knowledge graph so you can easily search and manage your code from within the Cline interface.

Why Cline?

Cline allows you to interface with powerful MCP servers right from your IDE. By connecting different servers and their “tools,” Cline can seamlessly enhance your development workflow - from generating code to analyzing complex structures - through natural language interactions.

How does Cognee integrate with Cline?

Cognee MCP server now provides code analysis capabilities. When integrated with Cline:

Cline sends task requests (e.g., code graph generation, code search) to Cognee.
Cognee uses its internal tools (like the codify tool) to analyze your repository, map code relationships, and build a knowledge graph.
Cline then displays the retrieved results - such as a code map or relevant snippets - right in your IDE, enabling you to navigate and manage your complex codebase more effectively.

Prerequisites

Before you proceed, make sure you have the following:

Visual Studio Code with the Cline extension installed.
A repository you wish to generate a knowledge graph from.
The Cognee repository cloned locally.
An LLM API key.

Do you have any specific environment constraints (e.g., Windows, macOS, Linux) or dependency management tools we should document? If so, let us know, and we’ll tailor instructions accordingly.

Step-by-Step Integration Guide

1. Install Cline

Install the Cline extension in VS Code:

2. Install and Start the Cognee Server

2.1 Clone the Cognee Repository

git clone https://www.github.com/topoteretes/cognee

Make sure you have uv.

brew install uv

And navigate to the cognee-mcp folder.

cd cognee/cognee-mcp

2.2 Install Dependencies

Once in the cognee-mcp directory, run:

uv sync --reinstall

2.3 Activate the Virtual Environment

source .venv/bin/activate

3. Configure Cline to Use Cognee

Cline requires a settings file that tells it where to find the Cognee server.

Open your Cline MCP settings file. The default location is:

~/Library/Application\ Support/CODE/user/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json

Edit the file (using your preferred text editor, e.g., nano) to add a new server entry for Cognee:

{
    "mcpServers": {
    "cognee": {
        "command": "uv",
        "args": [
        "--directory",
        "/{cognee_root_path}/cognee-mcp",
        "run",
        "cognee"
        ]
    }
    },
    "env": {
    "ENV": "local",
    "TOKENIZERS_PARALLELISM": "false",
    "LLM_API_KEY": "sk-..."
    }
}

/ {cognee_root_path}/cognee-mcp: Replace with the absolute path to where you cloned Cognee.
LLM_API_KEY: Replace the placeholder with your AI provider key.

Save the file.

4. Restart Cline

Once the configuration is saved, restart Cline (VS Code).
When Cline initializes, it will detect the new Cognee server from your config file.

5. Example - Use Cognee from Cline

Navigate to the repo you want to analyze within VS Code.
Open the Cline interface.
Type your natural language command to invoke the Cognee server.

With Cognee integrated into Cline, you can now seamlessly:

Generate detailed knowledge graphs of your complex codebases.
Search and manage tangled dependencies directly through natural language commands in your IDE.
Build more advanced workflows around code analysis, refactoring, or knowledge retrieval.

Join the Conversation!

Have questions? Join our community now to connect with professionals, share insights, and get your questions answered!

Promptfoo Continue