Cortex FAQs

Question 1

How does Cortex ensure privacy and security of my codebase?

Accepted Answer

Cortex is designed as a local-first tool. All your code and context data are stored entirely on your local machine under a `.context/` directory. No source code or indexed data is ever uploaded to the cloud, ensuring complete privacy.

Question 2

What is Cortex and how does it help developers?

Accepted Answer

Cortex is a local, repo-scoped context platform that indexes your codebase into structured entities (files, rules, ADRs). It provides AI coding assistants with semantic search, entity relationship graphs, and architectural context directly from your repository.

Question 3

What kind of advanced features does Cortex offer?

Accepted Answer

Beyond core semantic search and graph relationships, Cortex can experimentally extract function-level call graphs, perform impact analysis to show what's affected by a function change, and trace function calls, especially for JavaScript/TypeScript projects.

Question 4

Which AI coding assistants are compatible with Cortex?

Accepted Answer

Cortex is built to integrate seamlessly with AI coding assistants that support MCP (JSON-RPC over stdio), including Claude Code/Desktop and Codex MCP clients. It exposes its rich codebase context directly to these tools.

Question 5

How does Cortex keep my codebase context up-to-date?

Accepted Answer

Cortex features incremental updates and can run a background sync (watcher) to automatically keep your indexed codebase context fresh as changes occur in your repository. You can also manually trigger updates.

Cortex

Cortex

Key Features

Use Cases

Key Features

Use Cases